Advertisement

Utilization Trends and Factors Associated With ROS1 Testing Among Patients With Advanced Non–small-cell Lung Cancer in US Community Practices

Open AccessPublished:June 29, 2020DOI:https://doi.org/10.1016/j.cllc.2020.06.019

      Abstract

      Background

      Targeted therapy for patients with non–small-cell lung cancer (NSCLC) harboring ROS proto-oncogene 1 (ROS1) rearrangements was approved in 2016. However, little is known about real-world ROS1 testing practices in United States community practice. We aimed to characterize ROS1 testing rates and identify potential barriers to ROS1 testing.

      Patients and Methods

      Flatiron Health’s de-identified electronic health record-derived database was used to identify patients diagnosed with advanced NSCLC from July 2016 through December 2018 who received systemic treatment in a community practice setting. ROS1 and other biomarker testing was recorded. Regression analysis identified demographic and clinical characteristics associated with occurrence of ROS1 testing, longer time (≥ 25 days) from diagnosis to ROS1 result, and initiation of therapy prior to ROS1 result.

      Results

      Among 11,409 patients, documented ROS1 testing rates increased during the study period in squamous (from 30% to 48%) and nonsquamous (63% to 78%) histologies. Patients who were older, male, black, or with squamous histology, higher Eastern Cooperative Oncology Group score, recurrent disease, or history of smoking were significantly less likely to be tested. Among patients not tested for ROS1, 63% were tested for other biomarkers. Use of next-generation sequencing, older age, Hispanic/Latino ethnicity, squamous histology, de novo disease, and smoking history predicted longer time to test result post-diagnosis. Patients with delayed results were 9.7 times more likely to receive treatment prior to ROS1 test result.

      Conclusion

      In real-world practice, some patient subgroups may be less likely to receive timely ROS1 testing and to be identified for potential targeted therapy.

      Keywords

      Introduction

      In certain instances of non–small-cell lung cancer (NSCLC), the presence of molecular biomarkers can indicate the cancer may be susceptible to targeted therapy.
      • Bubendorf L.
      • Büttner R.
      • Al-Dayel F.
      • et al.
      Testing for ROS1 in non-small cell lung cancer: a review with recommendations.
      Among these, ROS proto-oncogene 1 (ROS1) rearrangement is a rare biomarker that occurs in approximately 1% to 2% of patients with NSCLC.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      ROS1 belongs to the human receptor tyrosine kinase family, and tyrosine kinase inhibitors have been investigated for treatment of NSCLC with ROS1 rearrangements.
      • Bubendorf L.
      • Büttner R.
      • Al-Dayel F.
      • et al.
      Testing for ROS1 in non-small cell lung cancer: a review with recommendations.
      For patients with ROS1-rearranged NSCLC, the guidelines of the National Comprehensive Cancer Network (NCCN) now recommend first-line therapy with the kinase inhibitors crizotinib
      (preferred), entrectinib
      (preferred), or ceritinib,
      and subsequent therapy with lorlatinib
      after disease progression.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      In March 2016, crizotinib was the first drug approved for treatment of patients with metastatic ROS1+ NSCLC by the United States (US) Food and Drug Administration,
      United States Food and Drug Administration
      FDA expands use of Xalkori to treat rare form of advanced non-small cell lung cancer [news release]. 2016.
      and this agent has been found effective for ROS1+ NSCLC in multiple clinical trials.
      • Shaw A.T.
      • Ou S.H.
      • Bang Y.J.
      • et al.
      Crizotinib in ROS1-rearranged non-small-cell lung cancer.
      • Mazières J.
      • Zalcman G.
      • Crino L.
      • et al.
      Crizotinib therapy for advanced lung adenocarcinoma and a ROS1 rearrangement: results from the EUROS1 cohort.
      • Gambacorti-Passerini C.
      • Orlov S.
      • Zhang L.
      • et al.
      Long-term effects of crizotinib in ALK-positive tumors (excluding NSCLC): a phase 1b open-label study.
      • Wu Y.L.
      • Yang J.C.
      • Kim D.W.
      • et al.
      Phase II study of crizotinib in East Asian patients with ROS1-positive advanced non-small-cell lung cancer.
      Entrectinib was approved in 2019 for treatment of metastatic ROS1+ NSCLC based on phase I and II trials.
      • Drilon A.
      • Siena S.
      • Ou S.I.
      • et al.
      Safety and antitumor activity of the multitargeted pan-TRK, ROS1, and ALK inhibitor entrectinib: combined results from two phase I trials (ALKA-372-001 and STARTRK-1).
      Additional trials have supported the potential efficacy of ceritinib,
      • Lim S.M.
      • Kim H.R.
      • Lee J.S.
      • et al.
      Open-label, multicenter, phase II study of ceritinib in patients with non-small-cell lung cancer harboring ROS1 rearrangement.
      lorlatinib,
      • Shaw A.T.
      • Felip E.
      • Bauer T.M.
      • et al.
      Lorlatinib in non-small-cell lung cancer with ALK or ROS1 rearrangement: an international, multicentre, open-label, single-arm first-in-man phase 1 trial.
      ,
      • Solomon B.J.
      • Besse B.
      • Bauer T.M.
      • et al.
      Lorlatinib in patients with ALK-positive non-small-cell lung cancer: results from a global phase 2 study.
      and cabozantinib
      • Hellerstedt B.A.
      • Vogelzang N.J.
      • Kluger H.M.
      • et al.
      Results of a phase II placebo-controlled randomized discontinuation trial of cabozantinib in patients with non-small-cell lung carcinoma.
      ,
      for NSCLC with ROS1 rearrangements, although as of this writing, the US Food and Drug Administration has not yet approved these medications for treatment of ROS1+ NSCLC.
      Because of the effectiveness of kinase inhibitors for treatment of NSCLC with ROS1 rearrangements, guidelines recommend testing for ROS1 rearrangement in most patients with advanced NSCLC. In 2018, updated molecular testing guidelines of the College of American Pathologists (CAP), the International Association for the Study of Lung Cancer (IASLC), and the Association for Molecular Pathology (AMP),
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      which are endorsed by the American Society of Clinical Oncology,
      • Kalemkerian G.P.
      • Narula N.
      • Kennedy E.B.
      • et al.
      Molecular testing guideline for the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors: American Society of Clinical Oncology endorsement of the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology clinical practice guideline update.
      added a strong recommendation for ROS1 testing for all patients with advanced-stage lung adenocarcinoma, regardless of clinical characteristics.
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      This testing may also be considered for histologies other than adenocarcinoma, depending on clinical features.
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      NCCN guidelines recommend testing for ROS1 in patients with advanced or metastatic NSCLC with nonsquamous histology and, in some cases, in patients with metastatic squamous NSCLC.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      Various methodologies are available for ROS1 testing, including fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), next-generation sequencing (NGS), and polymerase chain reaction (PCR); however, IHC testing for ROS1 has low specificity, and follow-up confirmatory testing is needed.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      ,
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      To minimize tissue use, ROS1 testing can be part of broad molecular profiling, including tests for anaplastic lymphoma kinase (ALK) rearrangements, epidermal growth factor receptor (EGFR) mutations, B-Raf proto-oncogene (BRAF) mutations, and neurotrophic receptor tyrosine kinase gene fusion.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      ,
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      ,
      • Kalemkerian G.P.
      • Narula N.
      • Kennedy E.B.
      • et al.
      Molecular testing guideline for the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors: American Society of Clinical Oncology endorsement of the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology clinical practice guideline update.
      As additional targeted therapies with efficacy for ROS1+ NSCLC become available, ensuring patients are tested for ROS1 in a timely manner is increasingly important so appropriate treatment can be determined. Currently information is limited about real-world ROS1 testing practices in the US.
      • Audibert C.M.
      • Shea M.B.
      • Glass D.J.
      • et al.
      Trends in the molecular diagnosis of lung cancer: results from an online market research survey. 2017.
      The objectives of this study were to describe current ROS1 testing rates in community oncology practices in the US treating patients with NSCLC, to identify patient characteristics or potential barriers that may be associated with lower rates of ROS1 testing, and to estimate the real-world timing of ROS1 testing and its potential impact on therapy decisions.

      Materials and Methods

      This study was a retrospective database analysis utilizing Flatiron Health’s nationwide, longitudinal, demographically and geographically diverse database,
      Flatiron Life Sciences
      Flatiron Health EHR-derived database. July 2019.
      which is derived from de-identified electronic health record (EHR) data from more than 280 cancer clinics and includes more than 2.2 million US patients with cancer available for analysis (April 2019 data cut). Copernicus Group Institutional Review Board approved the study protocol before study conduct; this approval included a waiver of informed consent.

      Patients

      Included patients were diagnosed with advanced (stage IIIB, IIIC, IV) NSCLC per the American Joint Committee on Cancer Staging Manual, seventh
      or eighth
      edition, as entered in the EHR by the treating clinician, between July 1, 2016, and December 31, 2018. Patients received first-line systemic treatment for NSCLC within 120 days of the advanced diagnosis. This inclusion criterion was intended to mitigate the risk of incomplete patient information (ie, after diagnosis, patient initiated treatment or was tested at a clinic not included in the database). Patients must have been treated in a community oncology practice setting. The study period (period for data extraction) extended from July 1, 2016, through March 31, 2019.
      Patient demographic variables included age, gender, race, ethnicity, insurance type (commercial, Medicare, Medicaid, or other/unknown), and region. Clinical measures included the patient’s score on the Eastern Cooperative Oncology Group (ECOG) Scale of Performance Status,
      ECOG-ACRIN Cancer Research Group
      ECOG performance status. 2019.
      histology of the patient’s NSCLC (squamous cell carcinoma, nonsquamous cell carcinoma, or not otherwise specified [NOS]), occurrence of the NSCLC (de novo or recurrent), and history of smoking. Weight loss was assessed using weight measures dated within 3 months before and 1 month after advanced diagnosis.

      ROS1 and Other Biomarker Testing

      ROS1 testing (if performed) was recorded for each patient. The type of ROS1 testing was recorded as FISH, IHC, NGS, PCR, other, or unknown. Testing for the following additional biomarkers was also recorded: programmed death-ligand 1 (PD-L1), EGFR, ALK, and BRAF. The Flatiron database documents testing reported in each patient’s EHR; testing performed at a clinic outside of the source EHR system may not be documented in the database.

      Timing of Key Outcomes

      Dates and relative timing of the following outcomes were available for each patient: diagnosis of advanced NSCLC, laboratory receipt of sample for testing, availability of ROS1 test results, and initiation of therapy.

      Statistical Analysis

      Descriptive statistics including means, medians, and proportions were used to summarize the patient population, biomarker testing rates, and timing of testing. Comparisons of baseline characteristics between patients receiving or not receiving ROS1 testing were performed using χ2 tests.
      Logistic and multinomial regression analyses were used to identify factors (demographic and clinical characteristics) associated with the following bivariate outcomes: ROS1 testing (or no ROS1 testing), longer times from advanced diagnosis to ROS1 test results (vs. shorter times), and initiation of therapy prior to ROS1 testing. Likelihood of an outcome given a certain predictor was expressed as an odds ratio (OR) with a 95% confidence interval (CI).

      Results

      Patients

      Among 54,326 patients in the database, 17,891 received a diagnosis of advanced NSCLC between July 1, 2016, and December 31, 2018 (see Supplemental Figure 1 in the online version). Of these patients, 11,409 received first-line therapy in a community practice setting; these patients comprised the study sample. Of the 11,409 total patients, 7871 (69.0%) had nonsquamous cell carcinoma, 3007 (26.4%) had squamous cell carcinoma, and 531 (4.7%) had NSCLC with histology NOS (Table 1). Of the 11,409 patients, 7616 (67%) received ROS1 testing and 3793 (33%) did not.
      Table 1Characteristics of Patients With NSCLC Who Did or Did Not Receive ROS1 Testing
      Characteristic, n (%)Not ROS1 Tested (n = 3793)ROS1 Tested (n = 7616)P Value
      Age, y<.0001
       <55254 (6.7)700 (9.2)
       55-64911 (24.0)1913 (25.1)
       65-741371 (36.1)2669 (35.0)
       ≥751257 (33.1)2334 (30.6)
      Female gender1558 (41.1)3646 (47.9)<.0001
      Race.0033
       Asian97 (2.6)236 (3.1)
       Black or African American362 (9.5)622 (8.2)
       Other331 (8.7)794 (10.4)
       Unknown442 (11.7)868 (11.4)
       White2561 (67.5)5096 (66.9)
      Hispanic/Latino ethnicity132 (3.5)263 (3.5).9411
      Payer.0047
       Commercial1737 (45.8)3668 (48.2)
       Medicare839 (22.1)1484 (19.5)
       Medicaid61 (1.6)92 (1.2)
       Other753 (19.9)1544 (20.3)
       Unknown403 (10.6)828 (10.9)
      Region.0006
       Midwest639 (16.8)1299 (17.1)
       Northeast787 (20.7)1616 (21.2)
       South1745 (46.0)3390 (44.5)
       West566 (14.9)1255 (16.5)
       Unknown56 (1.5)56 (0.7)
      Weight loss.2929
       >10% body weight164 (4.3)334 (4.4)
       <10% body weight
      Category for “< 10% weight loss” includes patients who gained weight or whose body weight was unchanged.
      2471 (65.1)5065 (66.5)
       Unknown1158 (30.5)2217 (29.1)
      ECOG PS score<.0001
       0-11797 (47.4)4118 (54.1)
       ≥2512 (13.5)876 (11.5)
       Unknown1484 (39.1)2622 (34.4)
      Histology<.0001
       NSCLC, histology NOS162 (4.3)369 (4.8)
       Nonsquamous cell carcinoma1909 (50.3)5962 (78.3)
       Squamous cell carcinoma1722 (45.4)1285 (16.9)
      Occurrence<.0001
       De novo2843 (75.0)6111 (80.2)
       Recurrent861 (22.7)1372 (18.0)
       Unknown89 (2.4)133 (1.7)
      Smoking status<.0001
       History of smoking3453 (91.0)6505 (85.4)
       No history of smoking329 (8.7)1103 (14.5)
       Unknown11 (0.3)8 (0.1)
      Year of advanced diagnosis<.0001
       2016
      In 2016, patients diagnosed from July 1 to December 31 were included.
      978 (25.8)1317 (17.3)
       20171573 (41.5)3206 (42.1)
       20181242 (32.7)3093 (40.6)
      Abbreviations: ECOG = Eastern Cooperative Oncology Group; NOS = not otherwise specified; NSCLC = non–small-cell lung cancer; PS = performance status
      a Category for “< 10% weight loss” includes patients who gained weight or whose body weight was unchanged.
      b In 2016, patients diagnosed from July 1 to December 31 were included.
      Characteristics of patients who did or did not receive ROS1 testing were generally similar (Table 1), except that patients who received ROS1 testing were more often female, had nonsquamous histology, had low ECOG scores (0 or 1), and had no history of smoking.

      ROS1 and Other Biomarker Testing

      Rates of documented ROS1 testing increased over the study period for patients with squamous histology (from 30% in quarter [Q]3 2016 to 48% in Q4 2018; 43% with all years combined) and nonsquamous histology (63% to 78%; 76% overall) (Figure 1). For patients with squamous histology, the rate of ROS1 testing was highest (57%) in the central South of the US, including Texas, and lowest (21%) in New England. Among patients with nonsquamous histology, rates of ROS1 testing were similar across US regions, ranging from 72% to 80%.
      Figure thumbnail gr1
      Figure 1Percent of Patients With NSCLC With Nonsquamous (n = 7871) or Squamous Cell Carcinoma (n = 3007) Who Received ROS1 Testing Over Time From Q3 2016 to Q4 2018
      Abbreviations: NSCLC = non–small-cell lung cancer; Q = quarter.
      FISH was the most common test type across all ROS1 tests (Figure 2), although the use of NGS testing increased during the course of the study relative to other ROS1 test types. Of 7616 patients tested for ROS1, 1470 (19%) patients had ≥ 2 ROS1 tests; the most common testing sequence was FISH followed by NGS (32% of patients with ≥ 2 tests). Of the 1470 patients with ≥ 2 ROS1 tests, 221 (15%) had an unsuccessful or indeterminate first test. Among 609 patients with ≥ 2 ROS1 tests and FISH as the first test, agreement between the first and second test types was high (> 97% for each test; 99.5% for FISH and NGS).
      Figure thumbnail gr2
      Figure 2Type of ROS1 Testing Received Over Time From Q3 2016 to Q4 2018 by Patients With NSCLC (n = 7616). Patients Could Receive More than One ROS1 Test; all ROS1 Tests Are Shown
      Abbreviations: FISH = fluorescence in situ hybridization; IHC = immunohistochemistry; NGS = next-generation sequencing; NSCLC = non–small-cell lung cancer; PCR = polymerase chain reaction; Q = quarter.
      Among 7616 patients tested for ROS1, most were also tested for 4 other key biomarkers (PD-L1, EGFR, ALK, and BRAF), regardless of histology of their NSCLC (see Supplemental Figure 2A in the online version). For example, > 97% were tested for ALK and EGFR. Among 3793 patients not tested for ROS1, 2381 (63%) were tested for other biomarkers, including ≥ 50% of patients with nonsquamous histology who had documented tests for each of PD-L1, EGFR, or ALK, and 52% of patients with squamous histology who were tested for PD-L1 (see Supplemental Figure 2B in the online version). Of 11,409 patients with NSCLC who received first-line systemic therapy, 1412 (12%) were not tested for biomarkers.

      Predictors of Receiving ROS1 Testing

      In the regression analysis, factors associated with significantly lower odds of being tested for ROS1 included age ≥ 75 years, male gender, black race, higher ECOG score, squamous or NOS histology, recurrent disease, history of smoking, and earlier year of diagnosis (Table 2). Having Medicaid or Medicare (vs. commercial) insurance was associated with lower likelihood of ROS1 testing, although the difference was not statistically significant.
      Table 2Factors Predicting ROS1 Testing Among Patients With NSCLC (N = 11,409)
      CharacteristicOR (95% CI)
      Age, y (ref: < 55)
       55-640.87 (0.74-1.04)
       65-740.90 (0.76-1.07)
       ≥750.82 (0.69-0.97)
      Female gender (ref: male)1.11 (1.02-1.21)
      Race (ref: white)
       Asian0.86 (0.66-1.11)
       Black or African American0.85 (0.73-0.99)
       Other1.14 (0.98-1.32)
       Unknown0.91 (0.79-1.04)
      Hispanic/Latino ethnicity (ref: not Hispanic/Latino)0.97 (0.76-1.23)
      Payer (ref: commercial)
       Medicare0.90 (0.80-1.00)
       Medicaid0.72 (0.50-1.02)
       Other1.02 (0.91-1.14)
       Unknown1.01 (0.88-1.17)
      Region (ref: West)
       Midwest0.98 (0.85-1.14)
       Northeast0.92 (0.80-1.07)
       South0.95 (0.83-1.07)
       Unknown0.39 (0.25-0.59)
      Weight loss (ref: < 10%)
       >10% body weight1.13 (0.92-1.39)
       Unknown0.94 (0.86-1.04)
      ECOG PS score (ref: 0-1)
       ≥20.78 (0.69-0.89)
       Unknown0.77 (0.70-0.85)
      Histology (ref: nonsquamous)
       NSCLC, histology NOS0.74 (0.61-0.89)
       Squamous cell carcinoma0.24 (0.22-0.27)
      Occurrence (ref: de novo)
       Recurrent0.79 (0.71-0.88)
       Unknown0.80 (0.60-1.08)
      History of smoking (ref: no history)0.77 (0.67-0.89)
      Year of advanced diagnosis (ref: 2016)
       20171.63 (1.46-1.82)
       20181.99 (1.78-2.23)
      Abbreviations: CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; NOS = not otherwise specified; NSCLC = non–small-cell lung cancer; OR = odds ratio; PS = performance status; ref = reference group

      Timing of ROS1 Test Result

      Large variability was observed in the time from diagnosis of advanced NSCLC to the ROS1 test result. Time from diagnosis to test result varied by test type, with shorter mean (SD) lags of 21 (13) days for IHC and 34 (50) days for FISH, compared with 47 (85) days for PCR and 64 (83) days for NGS. From laboratory receipt of the sample to ROS1 test result, the mean (SD) wait was 15 (27) days for IHC, 13 (11) days for NGS, 10 (14) days for FISH, and 4 (7) days for PCR.
      The strongest predictor of a longer lag between diagnosis and ROS1 test result was NGS (vs. FISH) test type (OR, 3.78; 95% CI, 3.28-4.34 for lag of 25-40 days; OR, 6.79; 95% CI, 5.86-7.85 for lag of ≥ 41 days) (Table 3). Additional significant predictors of a longer lag (25-40 or ≥ 41 vs. < 25 days) included age ≥ 65 years, Hispanic or Latino ethnicity, treatment in the US West region, squamous histology, de novo disease, history of smoking, and earlier diagnosis (in 2016) (Table 3).
      Table 3Factors Predicting Longer Time From Diagnosis of Advanced NSCLC to ROS1 Test Results (n = 7616)
      Characteristic25-40 Days vs. < 25 Days, OR (95% CI)≥ 41 Days vs. < 25 Days, OR (95% CI)
      Type of first ROS1 test (ref: FISH)
       IHC0.69 (0.36-1.32)0.37 (0.14-0.96)
       NGS3.78 (3.28-4.34)6.79 (5.86-7.85)
       Other/unknown1.82 (1.47-2.26)2.58 (2.07-3.21)
       PCR0.51 (0.37-0.70)1.28 (0.98-1.67)
      Age, y (ref: < 55)
       55-641.08 (0.85-1.36)1.09 (0.86-1.39)
       65-741.33 (1.05-1.67)1.25 (0.99-1.58)
       ≥751.38 (1.09-1.74)1.00 (0.78-1.27)
      Female gender (ref: male)1.01 (0.90-1.14)0.97 (0.86-1.1)
      Race (ref: White)
       Asian1.09 (0.77-1.53)0.70 (0.47-1.06)
       Black or African American0.95 (0.75-1.2)1.12 (0.89-1.41)
       Other1.02 (0.83-1.25)0.78 (0.62-0.97)
       Unknown1.06 (0.87-1.28)1.04 (0.85-1.28)
      Hispanic/Latino ethnicity (ref: not Hispanic/Latino)0.94 (0.66-1.35)1.48 (1.05-2.09)
      Payer (ref: commercial)
       Medicare1.05 (0.89-1.23)1.15 (0.97-1.36)
       Medicaid0.66 (0.35-1.24)1.10 (0.63-1.91)
       Other1.09 (0.93-1.28)1.14 (0.97-1.35)
       Unknown1.03 (0.84-1.27)1.18 (0.96-1.45)
      Region (ref: West)
       Midwest0.80 (0.65-0.98)0.61 (0.49-0.77)
       Northeast0.80 (0.66-0.98)0.82 (0.67-1.02)
       South0.75 (0.62-0.89)0.74 (0.61-0.89)
       Unknown0.68 (0.32-1.43)0.65 (0.31-1.36)
      Weight loss (ref: < 10%)
       >10% body weight0.81 (0.60-1.10)0.61 (0.43-0.87)
       Unknown1.85 (1.61-2.13)3.37 (2.92-3.88)
      ECOG PS score (ref: 0-1)
       ≥21.06 (0.88-1.28)0.88 (0.71-1.09)
       Unknown1.04 (0.90-1.19)1.27 (1.10-1.46)
      Histology (ref: nonsquamous)
       NSCLC, histology NOS1.18 (0.89-1.56)1.47 (1.10-1.95)
       Squamous cell carcinoma1.56 (1.32-1.85)2.31 (1.96-2.73)
      Occurrence (ref: de novo)
       Recurrent0.46 (0.39-0.55)0.58 (0.49-0.68)
       Unknown0.48 (0.30-0.78)0.54 (0.33-0.89)
      History of smoking (ref: no history)1.01 (0.85-1.21)1.32 (1.09-1.60)
      Year of advanced diagnosis (ref: 2016)
       20170.99 (0.82-1.18)0.60 (0.50-0.71)
       20180.94 (0.78-1.12)0.41 (0.34-0.48)
      Abbreviations: CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; FISH = fluorescence in situ hybridization; IHC = immunohistochemistry; NGS = next-generation sequencing; NOS = not otherwise specified; NSCLC = non–small-cell lung cancer; OR = odds ratio; PCR = polymerase chain reaction; PS = performance status; ref = reference group

      Initiation of Therapy Relative to ROS1 Test Result

      Among NSCLC patients with nonsquamous histology who were tested for ROS1, most initiated therapy after ROS1 test results were available (Figure 3). Among these patients, the proportion initiating therapy before receipt of test results declined from 40% in Q3 2016 to 26% in Q4 2018. A larger proportion of patients with squamous histology initiated treatment before ROS1 test results were received, but this proportion also declined from 66% to 36% over the study period. For both histologies, the proportion initiating therapy before ROS1 results stabilized at about 25% to 35% among those diagnosed with advanced disease in 2018 (Figure 3).
      Figure thumbnail gr3
      Figure 3Percent of NSCLC Patients Initiating Therapy before Receiving ROS1 Test Result Over Time From Q3 2016 to Q4 2018, by Histology and Year/Q of Advanced Diagnosis
      Abbreviation: NSCLC = non–small-cell lung cancer; Q = quarter.
      Patients whose ROS1 test results were delayed at least 25 days after diagnosis were almost 10 times more likely (OR, 9.73; 95% CI, 8.49-11.17) to initiate treatment before test results were received (Table 4). Patients whose first ROS1 test was NGS or PCR (vs. FISH) were more likely to initiate treatment before test results, whereas patients tested with IHC were less likely. In addition, patients with weight loss, squamous histology, or who were diagnosed in 2016 were more likely to initiate therapy before test results (Table 4).
      Table 4Factors Predicting Initiation of Treatment Before Receiving ROS1 Test Results (n = 7616)
      CharacteristicOR (95% CI)
      ≥25 days between advanced diagnosis and ROS1 result (ref: < 25 days)9.73 (8.49-11.17)
      Type of first ROS1 test (ref: FISH)
       IHC0.30 (0.11-0.82)
       NGS2.14 (1.87-2.44)
       Other/unknown1.38 (1.12-1.71)
       PCR1.37 (1.05-1.78)
      Age, y (ref: < 55)
       55–640.88 (0.70-1.10)
       65–740.83 (0.67-1.04)
       ≥750.61 (0.49-0.77)
      Female gender (ref: male)0.84 (0.74-0.95)
      Race (ref: white)
       Asian0.90 (0.62-1.29)
       Black or African American1.11 (0.89-1.38)
       Other0.92 (0.75-1.13)
       Unknown1.10 (0.91-1.33)
      Hispanic/Latino ethnicity (ref: not Hispanic/Latino)1.15 (0.83-1.59)
      Payer (ref: commercial)
       Medicare0.96 (0.82-1.13)
       Medicaid1.45 (0.85-2.45)
       Other1.09 (0.93-1.27)
       Unknown1.00 (0.82-1.22)
      Region (ref: West)
       Midwest0.82 (0.67-1.02)
       Northeast1.03 (0.85-1.25)
       South1.04 (0.87-1.24)
       Unknown0.97 (0.47-2.00)
      Weight loss (ref: < 10%)
       >10% body weight1.39 (1.04-1.85)
       Unknown0.27 (0.23-0.31)
      ECOG PS score (ref: 0–1)
       ≥20.89 (0.74-1.09)
       Unknown1.01 (0.88-1.15)
      Histology (ref: nonsquamous)
       NSCLC, histology NOS1.19 (0.91-1.55)
       Squamous cell carcinoma1.40 (1.20-1.63)
      Occurrence (ref: de novo)
       Recurrent0.68 (0.58-0.80)
       Unknown0.86 (0.54-1.37)
      History of smoking (ref: no history)1.12 (0.94-1.33)
      Year of advanced diagnosis (ref: 2016)
       20170.69 (0.59-0.82)
       20180.53 (0.45-0.63)
      Abbreviations: CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; FISH = fluorescence in situ hybridization; IHC = immunohistochemistry; NGS = next-generation sequencing; NOS = not otherwise specified; NSCLC = non–small-cell lung cancer; OR = odds ratio; PCR = polymerase chain reaction; PS = performance status; ref = reference group.

      Discussion

      In this retrospective analysis of biomarker testing patterns among 11,409 patients diagnosed with advanced NSCLC from 2016 through 2018, it was found that ROS1 testing rates have increased over time. For patients with nonsquamous histology, frequency of documented ROS1 testing plateaued at about 75% to 80% in 2018. Rates of ROS1 testing also continued to increase for patients with squamous histology. The lower rate of ROS1 testing for patients with squamous histology in this study appears to align with guidelines that strongly recommend ROS1 testing for advanced NSCLC with nonsquamous or adenocarcinoma histology but recommend this testing for only some squamous cases.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      ,
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      Despite a variety of test types being used to identify ROS1 mutations, concordance was high (> 97%) between tests, including between the most commonly used tests, FISH and NGS (99.5%), when more than 1 test was utilized.
      Previous studies have similarly shown an increasing rate of biomarker testing in the US over the past decade.
      • Pennell N.A.
      • Arcila M.E.
      • Gandara D.R.
      • West H.
      Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
      • Davies J.
      • Martinec M.
      • Coudert M.
      • Delmar P.
      • Crane G.
      Real-world anaplastic lymphoma kinase (ALK) rearrangement testing patterns, treatment sequences, and survival of ALK inhibitor-treated patients.
      • Enewold L.
      • Thomas A.
      Real-world patterns of EGFR testing and treatment with erlotinib for non-small cell lung cancer in the United States.
      • Gutierrez M.E.
      • Choi K.
      • Lanman R.B.
      • et al.
      Genomic profiling of advanced non-small cell lung cancer in community settings: gaps and opportunities.
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      • MacLean E.
      • Louder A.
      • Saverno K.
      • et al.
      Molecular testing patterns in metastatic non-small cell lung cancer.
      • Shen C.
      • Kehl K.L.
      • Zhao B.
      • et al.
      Utilization patterns and trends in epidermal growth factor receptor (EGFR) mutation testing among patients with newly diagnosed metastatic lung cancer.
      • Velcheti V.
      • Patwardhan P.D.
      • Liu F.X.
      • Chen X.
      • Cao X.
      • Burke T.
      Real-world PD-L1 testing and distribution of PD-L1 tumor expression by immunohistochemistry assay type among patients with metastatic non-small cell lung cancer in the United States.
      In a survey conducted in 2015, 38% of newly diagnosed patients with NSCLC, including 32% of those treated in a community-based center, were tested for ROS1,
      • Audibert C.M.
      • Shea M.B.
      • Glass D.J.
      • et al.
      Trends in the molecular diagnosis of lung cancer: results from an online market research survey. 2017.
      lower than rates observed from 2016 to 2018 in this study of patients with advanced NSCLC.
      Among patients with both squamous and nonsquamous histologies who were not tested for ROS1, more than one-half were tested for PD-L1. This finding may indicate that PD-L1 testing is prioritized over ROS1 testing for many patients. CAP/IASLC/AMP guidelines published in 2018 did not make a recommendation regarding PD-L1 testing.
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      However, in 2019, NCCN guidelines elevated PD-L1 testing for patients with metastatic NSCLC to a category 1 level of recommendation, whereas ROS1 testing is a category 2A recommendation for patients with nonsquamous NSCLC.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      EGFR testing and ALK testing are category 1 recommendations for nonsquamous NSCLC.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      In the current study, PD-L1 testing was especially likely to be performed among patients with squamous histology who were not tested for ROS1, consistent with the NCCN recommendations.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      Given the association of ROS1-rearranged NSCLC with features that predict against benefit from immunotherapy,
      • Singal G.
      • Miller P.G.
      • Agarwala V.
      • et al.
      Association of patient characteristics and tumor genomics with clinical outcomes among patients with non-small cell lung cancer using a clinicogenomic database.
      it is concerning that physicians rely on the results of PD-L1 expression for therapeutic decision-making prior to receipt of results of molecular testing, including for ROS1 alterations.
      Certain patients appeared less likely than others to receive ROS1 testing, including patients who are older (≥ 75 years), male, or black, and patients with higher ECOG scores, squamous histology, recurrent disease, or history of smoking. Other studies have also found that younger patients,
      • Enewold L.
      • Thomas A.
      Real-world patterns of EGFR testing and treatment with erlotinib for non-small cell lung cancer in the United States.
      ,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      ,
      • Shen C.
      • Kehl K.L.
      • Zhao B.
      • et al.
      Utilization patterns and trends in epidermal growth factor receptor (EGFR) mutation testing among patients with newly diagnosed metastatic lung cancer.
      women,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      patients living in the West,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      ,
      • Shen C.
      • Kehl K.L.
      • Zhao B.
      • et al.
      Utilization patterns and trends in epidermal growth factor receptor (EGFR) mutation testing among patients with newly diagnosed metastatic lung cancer.
      patients with commercial insurance (vs. Medicaid or Medicare),
      • Enewold L.
      • Thomas A.
      Real-world patterns of EGFR testing and treatment with erlotinib for non-small cell lung cancer in the United States.
      ,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      patients with de novo (vs. recurrent) disease,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      and patients without history of smoking
      • Gutierrez M.E.
      • Choi K.
      • Lanman R.B.
      • et al.
      Genomic profiling of advanced non-small cell lung cancer in community settings: gaps and opportunities.
      ,
      • Illei P.B.
      • Wong W.
      • Wu N.
      • et al.
      ALK testing trends and patterns among community practices in the United States.
      were more likely to receive biomarker testing, although not all of these factors were predictive in all studies. These findings suggest an opportunity to provide additional education to clinicians in community practice on the guidelines recommendation that all patients with advanced-stage lung adenocarcinoma should be tested for ROS1, regardless of clinical characteristics
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      or type of health insurance.
      • Pennell N.A.
      • Arcila M.E.
      • Gandara D.R.
      • West H.
      Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
      Given the decision of the Centers for Medicare & Medicaid Services to cover NGS tests for patients with cancer
      • Pennell N.A.
      • Arcila M.E.
      • Gandara D.R.
      • West H.
      Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
      and the inability to predict the presence of an oncogenic driver with clinical features alone, these data suggest that biomarker testing is not likely being utilized as frequently as it should be for patients with lung cancer.
      Use of NGS testing for ROS1 has increased over time, and NGS was as commonly used as FISH by the end of 2018. The mean times between receipt of sample and ROS1 test results were 15 days or less for NGS, FISH, PCR, and IHC, falling close to the 10-day turnaround recommended in CAP/IASLC/AMP guidelines.
      • Lindeman N.I.
      • Cagle P.T.
      • Aisner D.L.
      • et al.
      Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
      However, based on time from diagnosis to receipt of ROS1 test results (as high as a mean of 64 days for NGS), initiation of ROS1 testing was delayed for many patients. Although the reasons for this delay are unknown and beyond the scope of this study, delays in ordering or batching/transportation of samples to distant laboratories or reimbursement challenges could be potential reasons for this observation.
      • Pennell N.A.
      • Arcila M.E.
      • Gandara D.R.
      • West H.
      Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
      Reflex testing on diagnosis (without requirement for a physician’s order) and ensuring that adequate tissue for biomarker testing is obtained at biopsy can improve rates and timeliness of testing.
      • Pennell N.A.
      • Arcila M.E.
      • Gandara D.R.
      • West H.
      Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
      The clinical implications of delays in testing are apparent, as longer time to receipt of ROS1 test results was a major predictor of initiation of therapy before test results were received. Overall, in mid to late 2018, therapy was initiated for ∼25% to 35% of patients before receipt of their ROS1 test results. The importance of receiving ROS1 results before initiating therapy is highlighted by 2019 NCCN guidelines, which advise crizotinib, entrectinib, or ceritinib as first-line therapy for ROS1 rearrangement–positive NSCLC.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      When ROS1 rearrangement is discovered during first-line therapy, these guidelines recommend either completing planned systemic therapy or interrupting that therapy to initiate crizotinib (preferred), entrectinib (preferred), or ceritinib.
      National Comprehensive Cancer Network
      NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
      Thus, timely awareness of ROS1 rearrangement can facilitate optimal therapy choices and avoid potentially unnecessary exposure to chemotherapy.
      Limitations of this study include those common to all real-world data sources, such as potential for data misclassification. For example, some patients may have been tested for ROS1 or other biomarkers by facilities outside the Flatiron network that did not record the tests in the Flatiron EHR database. Those patients would have been misclassified in the current analyses as not having been tested. The analysis was constrained by the limited number of status indicators, such as ECOG score and weight loss data, available in the database. A lack of information about the date when each test was ordered also limits the ability to interpret findings. Lastly, we could not obtain reasons for characteristics associated with testing and delays in testing from this database. Future qualitative research would be needed to understand the reasons behind these observations.

      Conclusion

      In conclusion, this analysis identified patient characteristics associated with a lower likelihood of ROS1 testing and testing delays. Some patient subgroups may therefore be less likely to be identified for potential targeted therapy in a timely manner. The study findings highlight the importance of ensuring timely biomarker testing for all patients diagnosed with advanced NSCLC.

      Clinical Practice Points

      • ROS1 rearrangement occurs in approximately 1% to 2% of patients with NSCLC. Targeted therapies have been approved for treatment of ROS1-rearranged NSCLC.
      • Because of the availability of targeted treatment, guidelines recommend testing for ROS1 rearrangement in most patients with advanced NSCLC. However, it is not known how often patients are tested in US community practice.
      • Using a large EHR-derived database, ROS1 testing patterns among 11,409 patients diagnosed with advanced NSCLC were analyzed.
      • Rates of ROS1 testing among patients with NSCLC with nonsquamous histology increased from 63% in 2016 to 78% in 2018.
      • However, patients with certain characteristics were less likely to be tested, including patients who were older, male, or black or had higher ECOG score, recurrent disease, or history of smoking.
      • Additionally, patients who were older, who were Hispanic or Latino, or who had a history of smoking were more likely to wait longer for ROS1 test results, and patients with delayed results were almost 10 times more likely to receive treatment before ROS1 test results.
      • These findings highlight the importance of implementing timely biomarker testing for all patients diagnosed with advanced NSCLC in community practice so that patients who may benefit from targeted therapy will be identified in a timely manner.

      Disclosure

      W. Wong declares employment by Genentech. N. Wu declares employment by Genentech at the time of the analysis. R. Gupta declares employment by Genentech. A.S. Mansfield declares research funding from Novartis and Verily (for Verily: commercial research grants paid to Dr Mansfield’s institution); an advisory role for AbbVie, Genentech, and Bristol Myers Squibb (for BMS: honoraria paid to Dr Mansfield’s institution); and a consulting/advisory role with AstraZeneca.

      Acknowledgments

      Editorial support for this article, including copyediting and production assistance, was provided by Lisa Baker, PhD, Melissa L. Bogen, ELS, and Esther Tazartes, MS, of the Global Outcomes Group. This paper, including medical writing support, was funded by Genentech, Inc, South San Francisco, CA.

      Supplemental Data

      Figure thumbnail fx1
      Supplemental Figure 1Selection of Patients for Inclusion in Study Analysis
      Abbreviation: NSCLC = non–small-cell lung cancer.
      Figure thumbnail fx2
      Supplemental Figure 2Frequency of Testing for Biomarkers Other than ROS1 Among Patients With NSCLC Who Were Tested for ROS1 (n = 7616) (A) or Not Tested for ROS1 (n = 3793) (B)
      Abbreviations: ALK = anaplastic lymphoma kinase; BRAF = B-Raf proto-oncogene; EGFR = epidermal growth factor receptor; NOS = not otherwise specified; NSCLC = non–small-cell lung cancer; PD-L1 = programmed death-ligand 1.

      References

        • Bubendorf L.
        • Büttner R.
        • Al-Dayel F.
        • et al.
        Testing for ROS1 in non-small cell lung cancer: a review with recommendations.
        Virchows Arch. 2016; 469: 489-503
        • National Comprehensive Cancer Network
        NCCN clinical practice guidelines in oncology (NCCN Guidelines): non-small cell lung cancer, version 7.2019. 2019.
        (Available at:)
      1. Xalkori [Prescribing Information]. Pfizer Inc, New York, NY2019
      2. Rozlytrek [Prescribing Information]. Genentech USA Inc, South San Francisco, CA2019
      3. Zykadia [Prescribing Information]. Novartis Pharmaceuticals Corporation, East Hanover, NJ2019
      4. Pfizer Inc. Lorbrena [Package Insert]. Pfizer, Inc, New York, NY2018
        • United States Food and Drug Administration
        FDA expands use of Xalkori to treat rare form of advanced non-small cell lung cancer [news release]. 2016.
        (Available at:)
        • Shaw A.T.
        • Ou S.H.
        • Bang Y.J.
        • et al.
        Crizotinib in ROS1-rearranged non-small-cell lung cancer.
        N Engl J Med. 2014; 371: 1963-1971
        • Mazières J.
        • Zalcman G.
        • Crino L.
        • et al.
        Crizotinib therapy for advanced lung adenocarcinoma and a ROS1 rearrangement: results from the EUROS1 cohort.
        J Clin Oncol. 2015; 33: 992-999
        • Gambacorti-Passerini C.
        • Orlov S.
        • Zhang L.
        • et al.
        Long-term effects of crizotinib in ALK-positive tumors (excluding NSCLC): a phase 1b open-label study.
        Am J Hematol. 2018; 93: 607-614
        • Wu Y.L.
        • Yang J.C.
        • Kim D.W.
        • et al.
        Phase II study of crizotinib in East Asian patients with ROS1-positive advanced non-small-cell lung cancer.
        J Clin Oncol. 2018; 36: 1405-1411
        • Drilon A.
        • Siena S.
        • Ou S.I.
        • et al.
        Safety and antitumor activity of the multitargeted pan-TRK, ROS1, and ALK inhibitor entrectinib: combined results from two phase I trials (ALKA-372-001 and STARTRK-1).
        Cancer Discov. 2017; 7: 400-409
        • Lim S.M.
        • Kim H.R.
        • Lee J.S.
        • et al.
        Open-label, multicenter, phase II study of ceritinib in patients with non-small-cell lung cancer harboring ROS1 rearrangement.
        J Clin Oncol. 2017; 35: 2613-2618
        • Shaw A.T.
        • Felip E.
        • Bauer T.M.
        • et al.
        Lorlatinib in non-small-cell lung cancer with ALK or ROS1 rearrangement: an international, multicentre, open-label, single-arm first-in-man phase 1 trial.
        Lancet Oncol. 2017; 18: 1590-1599
        • Solomon B.J.
        • Besse B.
        • Bauer T.M.
        • et al.
        Lorlatinib in patients with ALK-positive non-small-cell lung cancer: results from a global phase 2 study.
        Lancet Oncol. 2018; 19: 1654-1667
        • Hellerstedt B.A.
        • Vogelzang N.J.
        • Kluger H.M.
        • et al.
        Results of a phase II placebo-controlled randomized discontinuation trial of cabozantinib in patients with non-small-cell lung carcinoma.
        Clin Lung Cancer. 2019; 20: 74-81.e1
      5. Cabometyx [Prescribing Information]. Exelixis, Alameda, CA2019
        • Lindeman N.I.
        • Cagle P.T.
        • Aisner D.L.
        • et al.
        Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology.
        Arch Pathol Lab Med. 2018; 142: 321-346
        • Kalemkerian G.P.
        • Narula N.
        • Kennedy E.B.
        • et al.
        Molecular testing guideline for the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors: American Society of Clinical Oncology endorsement of the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology clinical practice guideline update.
        J Clin Oncol. 2018; 36: 911-919
        • Audibert C.M.
        • Shea M.B.
        • Glass D.J.
        • et al.
        Trends in the molecular diagnosis of lung cancer: results from an online market research survey. 2017.
        (Available at:)
        • Flatiron Life Sciences
        Flatiron Health EHR-derived database. July 2019.
        (Available at:)
        https://flatiron.com/real-world-evidence/
        Date accessed: November 17, 2019
      6. Edge S. Byrd D.R. Compton C.C. AJCC Cancer Staging Manual. 7th ed. Springer International Publishing, Basel, Switzerland2011
      7. Amin M.B. Edge S.B. Greene F.L. AJCC Cancer Staging Manual. 8th ed. Springer International Publishing, Basel, Switzerland2017
        • ECOG-ACRIN Cancer Research Group
        ECOG performance status. 2019.
        (Available at:)
        • Pennell N.A.
        • Arcila M.E.
        • Gandara D.R.
        • West H.
        Biomarker testing for patients with advanced non-small cell lung cancer: real-world issues and tough choices.
        Am Soc Clin Oncol Educ Book. 2019; 39: 531-542
        • Davies J.
        • Martinec M.
        • Coudert M.
        • Delmar P.
        • Crane G.
        Real-world anaplastic lymphoma kinase (ALK) rearrangement testing patterns, treatment sequences, and survival of ALK inhibitor-treated patients.
        Curr Med Res Opin. 2019; 35: 535-542
        • Enewold L.
        • Thomas A.
        Real-world patterns of EGFR testing and treatment with erlotinib for non-small cell lung cancer in the United States.
        PLoS One. 2016; 11: e0156728
        • Gutierrez M.E.
        • Choi K.
        • Lanman R.B.
        • et al.
        Genomic profiling of advanced non-small cell lung cancer in community settings: gaps and opportunities.
        Clin Lung Cancer. 2017; 18: 651-659
        • Illei P.B.
        • Wong W.
        • Wu N.
        • et al.
        ALK testing trends and patterns among community practices in the United States.
        JCO Precis Oncol. 2018; 2: 1-11
        • MacLean E.
        • Louder A.
        • Saverno K.
        • et al.
        Molecular testing patterns in metastatic non-small cell lung cancer.
        Am J Manag Care. 2016; 22: e60-e67
        • Shen C.
        • Kehl K.L.
        • Zhao B.
        • et al.
        Utilization patterns and trends in epidermal growth factor receptor (EGFR) mutation testing among patients with newly diagnosed metastatic lung cancer.
        Clin Lung Cancer. 2017; 18: e233-e241
        • Velcheti V.
        • Patwardhan P.D.
        • Liu F.X.
        • Chen X.
        • Cao X.
        • Burke T.
        Real-world PD-L1 testing and distribution of PD-L1 tumor expression by immunohistochemistry assay type among patients with metastatic non-small cell lung cancer in the United States.
        PLoS One. 2018; 13: e0206370
        • Singal G.
        • Miller P.G.
        • Agarwala V.
        • et al.
        Association of patient characteristics and tumor genomics with clinical outcomes among patients with non-small cell lung cancer using a clinicogenomic database.
        JAMA. 2019; 321: 1391-1399