- Email: [email protected]
4B.01 Lung Cancer Patients Migrate to Seek Better Care
November 2017
Oral Sessions
or 6300 folds respectively. Results: Detection limit for mutant allele frequency in our study was 0.1%. The sequencing results were analyzed by bioinformatic expertise based on our previous studies on the baseline mutation profiling of circulating cell-free DNA and the clinicopathological data of these patients. Among all the 27 lung cancer patients, 80 percent were predicted as positive by ctDNA sequencing when the standard was defined as at least one of the hotspot mutations detected in the blood (ctDNA) was also detected in tumor tissue. Pneumonia and pulmonary tuberculosis were detected as negative according to the above standard. When evaluating all hotspots, 949 of 1265 (75 percent) mutations detected in tumor tissue were also detected in patients’ blood. When evaluating all genetic variations, including those present at high levels in tumor tissue (clonal, driver genes in the panel) as well as those at low levels (sub-clonal, passenger genes in the panel), 327 of 583 (56 percent) detected in tumor tissue were also detected in patients’ blood. Conclusion: We demonstrated the importance of sequencing both circulating cell-free DNA and genomic DNA in tumor tissue for ctDNA detection in lung cancer. We also determined and confirmed the consistency of ctDNA and tumor tissue through NGS according to the criteria explored in our studies. Our strategy can initially distinguish the lung cancer from other space occupying lesions of lung. Our work shows that the consistency will be benefited from the optimization in sensitivity and specificity in ctDNA detection.
S1559
consisting of adenocarcinoma (46%), squamous cell (42.5%) and NSCLC (not specified) in the rest. Most patients were Stage IV (70.7%), with the most common metastatic site being bones (35.4%) followed by liver (24.2%). 64.6% had at least 1 metastatic compartment involved at anti-PD-1 initiation. The median baseline CRP was 22 mg/L. 38.4% experienced irAEs during treatment. A significant increase in CRP was seen at the time of grade II-IV irAEs compared to baseline (P<0.001). After a median follow-up of 8.5 months, 50% patients were alive. Both univariate and multivariate Cox regression identified liver involvement and CRP> 50 to be associated with inferior OSI. Using log rank method, median OSI for CRP 50 was y 3.5 times that of CRP >50 (9.3 vs 2.7 months, P¼ 0.014). Conclusion: Baseline CRP cut off of 50 mg/L can serve as an independent predictive biomarker in lung cancer patients on anti-PD-1 therapy. Incorporation of CRP and its further validation for predicting survival outcomes in prospective lung cancer trials is required.
JUNIOR BREAKOUT SESSION: 4B POPULATION SCIENCE FRIDAY, SEPTEMBER 15, 2017 4B.01 Lung Cancer Patients Migrate to Seek Better Care Topic: Medical Oncology
4A.03 Predictive Utility of c-reactive Protein (CRP) in Advanced Stage Lung Cancer Treated with AntiProgrammed Cell Death-1 (PD-1) Therapy
D. Pham,1 C. Pinkston,2 M. Oechsli,1 M. Kong,2 J. Rios-Perez,1 G. Kloecker1 1James Graham Brown Cancer Center, University of Louisville, Louisville, KY/US, 2School Of Public Health & Information Sciences, University of Louisville, Louisville, KY/US
Topic: Medical Oncology 1
2
3
4
5
A.R. Naqash, C.R.G. Stroud, C. Cherry, M. Muzaffar, M. Bowling, P.R. Walker6 1Hematology/Oncology, East Carolina University/Vidant Medical Center, Greenville, NC/US, 2Hematology/Oncology, East Carolina University/Vidant Medical Center, Greenville, NC/US, 3Hematology Oncology, East Carolina University Brody School of Medicine, Greenville, NC/US, 4Hematology/Oncology, East Carolina University, Greenville, NC/US, 5Pulmonary Critical Care, East Carolina University Brody School of Medicine, Greenville, NC/US, 6Hematology/Oncology, Brody School of Medicine at East Carolina University, Greenville, NC/US Background: The identification of surrogate biomarkers with a predictive value in patients treated with anti-PD-1 therapy has been a challenge. CRP is an acute phase protein of hepatic origin secreted in response to pro-inflammatory cytokines. Various studies have elucidated an elevated CRP to correlate with inferior outcomes in a multitude of malignancies. We sought to evaluate the predictive utility of CRP in the context of lung cancer treated with anti-PD-1 therapy. Methods: We identified a total of 99 patients with advanced non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) between April 2015 to March 2017 treated with nivolumab after a first line platinum-doublet. Baseline CRP was measured at anti-PD-1 initiation as well as serially with subsequent doses. Cox regression was used to predict associations between survival and candidate inflammatory biomarkers (CRP, albulin) as well as clinical characteristics [age, gender, race, stage, number of metastatic compartments, presence of liver/ brain/bone metastasis, immune-related adverse events (irAEs)]. Overall survival after immunotherapy (OSI) was defined as time from immunotherapy initiation to last follow-up or death. Based on the webbased program designed by Budczies et al, optimal cut-off for CRP in relation to OSl was identified to be 49.1 (approximated to 50 mg/L). Kaplan-Meier method was used to assess survival outcomes. Results: Median age was 65 years with a majority being Caucasian (64.7%) and males (64.6%). NSCLC was predominant (88%), with histology
Background: Every year a significant population exists of those diagnosed with NSCLC who choose not to get treatment upon diagnosis and then “migrate” to additional hospital systems before ultimately getting treatment. Migration to different hospitals may play a role in the decision to treat or not-to-treat, and we aimed to evaluate the potential factors that lead to treatment. Methods: As part of the Kentucky LEADS Collaborative, 27 of 31 Kentucky hospital registries contacted provided NSCLC data of 5,807 out of 10,333 patients from 2012-2014. Variables collected included hospital accreditation status by the Commission on Cancer, patient zip codes, age at diagnosis, stage, overall survival (OS), sex, race, education, income, and insurance status. Treatment included any combination of surgery, radiation, or chemotherapy. Hospital records were matched to Kentucky Cancer Registry records to determine the number of hospitals visited for treatment. Patient treatment and migration patterns were analyzed with a logistic regression model along with additional post-hoc analysis. Results: Treatment, initially or via migration, was more likely when an accredited hospital was visited (84% vs. 59%). Most patients were treated at their initial hospital (73%). However, among the remaining patients, 37% migrated to a different hospital where most then received treatment (92%). Initially treated vs untreated was significantly associated with Stage I-II disease, insurance status, younger age (66.8 vs 70.1 years), and longer OS (506 days vs 306). Migrating to another hospital was associated with Stage IIII disease, younger age (66.4 vs 72.2 years), longer OS (561 vs 157 days), but also notably associated with initial hospitals missing treatment modalities and patients having private insurance. Patients who were treated after migrating were associated with Stage I-II disease, younger age (65.8 vs 72.8 years), and longer OS (595 vs 153 days). Compared to patients who were treated initially, patients treated via migration lived longer (595 vs 506 days) and particularly had longer survival with stage IV disease (414.2 vs 321.6 days). Conclusion: This analysis demonstrates a survival benefit for initially untreated patients who migrate to another hospital. This migration is significantly associated with stage, missing treatment modalities at diagnosis site, and
S1560 insurance status which suggests that patients intending to seek better care will frequently migrate. They are more likely to receive treatment and live longer if they are insured. Considering the current landscape of changing healthcare policy, it is notable that insurance status plays such a significant role in enabling lung cancer patients to find effective treatment.
Journal of Thoracic Oncology
Vol. 12 No. 11S1
4B.03 Prospective Evaluation of Multidisciplinary Lung Cancer Care: Timeliness, Thoroughness, and Patient/ Caregiver Perspectives Topic: Medical Oncology
4B.02 Increasing Survival in Stage IV NSCLC in Academic versus Community Based Centers in the National Cancer Database Topic: Medical Oncology S. Ramalingam,1 M. Dinan,1 J. Crawford2 1Duke University, Durham, NC/US, 2Medical Oncology, Duke University, Durham, NC/US Background: Overall survival of patients in the United States with advanced Non-small cell lung cancer (NSCLC) remains poor. However, improved outcomes are expected from recent advances in management from optimized chemotherapy and biomarker driven personalized therapies. With the increased early use of emerging regimens, we hypothesized that survival differences between academic and community centers are widening over time. Methods: We did a retrospective analysis of patients diagnosed with stage IV NSCLC between 1998 and 2010 within the 2012 National Cancer Database (NCDB). The primary end point was 2-year survival; analysis was limited to patients diagnosed up to 2010 to allow at least 2 years of follow-up. We restricted our analysis to hospitals present in the dataset throughout the study period. Overall survival was compared by academic versus community site of treatment controlling for the year of diagnosis, age, gender, histology, and insurance status. Results: In our multivariable analysis of 193,279 patients with stage IV NSCLC, 2-year survival differentially improved over time between academic and community-based centers (p¼0.0005) at a rate of 0.15% per year with 99.9% confidence interval (0.008%,0.3%). Patients treated in community centers tended to be older (mean 64 versus 62 years) and on average traveled 15 miles less for treatment. These patient differences between community and academic centers were stable over time. Conclusion: We found a widening gap in 2-year survival between academic and community centers in the NCDB. Further research is needed to understand and address widening disparities in outcomes between academic and community centers, and the potential impact of immune approaches on these outcomes.
M. Smeltzer,1 F.E. Rugless,2 H.K. Lee,3 K. Ward,1 N.R. Faris,2 M.A. Ray,1 M. Meadows,1 B. Jiang,2 B. Jackson,2 C. Foust,2 A. Patel,2 N. Boateng,2 S. Kedia,1 K. Roark,2 C. Houston-Harris,2 C. Fehnel,2 R.S. Signore,2 R. Fox,2 E.T. Robbins,2 J. Li,3 R.U. Osarogiagbon2 1 School Of Public Health, University of Memphis, Memphis, TN/US, 2 Baptist Cancer Center, Memphis, TN/US, 3Dept Of Industrial and Systems Engineering, University of Wisconsin-Madison, Madison, WI/US Background: The process of care for Non-small Cell Lung Cancer (NSCLC) patients provides many challenges. A Multidisciplinary (MD) model may improve outcomes, but its value has been difficult to quantify, with most studies being retrospective pre-post analyses. Timeliness of care is one MD advantage that studies consistently demonstrate, but more timely care does not necessarily translate to improved survival. However, timeliness is a patient-centered value, affirmed in our pre-study focus groups. We prospectively evaluated timeliness, staging activities, stage-based treatment selection, and patient/caregiver perspectives in MD vs. usual serial care (SC). Methods: This comparative effectiveness trial evaluated MD clinic vs. SC on multiple patient-centered endpoints with mixed-methods. Timeliness was evaluated by process engineering with bottleneck analyses. The 5-step process of care was: lesion identification, diagnostic biopsy, non-invasive staging, invasive staging, and definitive treatment. Quality of staging endpoints included bimodal staging (CT and PET or invasive biopsy) and trimodal staging (using all 3). Stage based treatments were compared with NCCN recommendations. Patient/caregiver satisfaction with quality of care and timeliness was measured by patient/caregiver surveys at baseline, 3-months, and 6months. Statistical methods: Chi-Square test, Wilcoxon-Mann-Whitney test. Results: The study enrolled 527 patients (178 MD, 349 SC). Average waiting times between steps were 48.8 days (MD) and 43.3 days (SC). Bottlenecks included time from: lesion identification to diagnostic biopsy, invasive or noninvasive staging to definitive treatment. MD patients received higher rates of bimodal staging (89% vs. 76%, p-value<0.001) and tri-modal staging (55% vs. 37%, pvalue<0.001) than SC patients. Ultimately, MD patients received more stage appropriate treatment than SC (79% vs. 68%, p-value<0.01). Overall, patients on MD and SC reported satisfaction with timeliness measures at baseline, 3-, and 6-months, scoring most items similarly high (between 2.7/3.0 and 2.9/3.0; individual question response rates 48-86%). At 6-months, MD patients were more satisfied with overall time from diagnosis to treatment completion (4.9/5.0 vs. 4.6/5.0, pvalue¼0.03). MD patients also reported higher satisfaction with combined quality of care from team members at all 3 time points (6Month: 15.9/16.0 vs. 14.4/16.0, p<0.0001). MD Caregivers were more satisfied with overall quality of care compared with other NSCLC patients at baseline (p<0.01) and 6-months (p<0.01). Conclusion: Despite the additional time to treatment initiation, MD patients received more thorough staging, more stage-appropriate treatment, and were more satisfied with the timeliness of their care. Timeliness is not an absolute quantity, but may be a relative value based on patients’ satisfaction with their level of engagement in the care-delivery process.
Oral Sessions
or 6300 folds respectively. Results: Detection limit for mutant allele frequency in our study was 0.1%. The sequencing results were analyzed by bioinformatic expertise based on our previous studies on the baseline mutation profiling of circulating cell-free DNA and the clinicopathological data of these patients. Among all the 27 lung cancer patients, 80 percent were predicted as positive by ctDNA sequencing when the standard was defined as at least one of the hotspot mutations detected in the blood (ctDNA) was also detected in tumor tissue. Pneumonia and pulmonary tuberculosis were detected as negative according to the above standard. When evaluating all hotspots, 949 of 1265 (75 percent) mutations detected in tumor tissue were also detected in patients’ blood. When evaluating all genetic variations, including those present at high levels in tumor tissue (clonal, driver genes in the panel) as well as those at low levels (sub-clonal, passenger genes in the panel), 327 of 583 (56 percent) detected in tumor tissue were also detected in patients’ blood. Conclusion: We demonstrated the importance of sequencing both circulating cell-free DNA and genomic DNA in tumor tissue for ctDNA detection in lung cancer. We also determined and confirmed the consistency of ctDNA and tumor tissue through NGS according to the criteria explored in our studies. Our strategy can initially distinguish the lung cancer from other space occupying lesions of lung. Our work shows that the consistency will be benefited from the optimization in sensitivity and specificity in ctDNA detection.
S1559
consisting of adenocarcinoma (46%), squamous cell (42.5%) and NSCLC (not specified) in the rest. Most patients were Stage IV (70.7%), with the most common metastatic site being bones (35.4%) followed by liver (24.2%). 64.6% had at least 1 metastatic compartment involved at anti-PD-1 initiation. The median baseline CRP was 22 mg/L. 38.4% experienced irAEs during treatment. A significant increase in CRP was seen at the time of grade II-IV irAEs compared to baseline (P<0.001). After a median follow-up of 8.5 months, 50% patients were alive. Both univariate and multivariate Cox regression identified liver involvement and CRP> 50 to be associated with inferior OSI. Using log rank method, median OSI for CRP 50 was y 3.5 times that of CRP >50 (9.3 vs 2.7 months, P¼ 0.014). Conclusion: Baseline CRP cut off of 50 mg/L can serve as an independent predictive biomarker in lung cancer patients on anti-PD-1 therapy. Incorporation of CRP and its further validation for predicting survival outcomes in prospective lung cancer trials is required.
JUNIOR BREAKOUT SESSION: 4B POPULATION SCIENCE FRIDAY, SEPTEMBER 15, 2017 4B.01 Lung Cancer Patients Migrate to Seek Better Care Topic: Medical Oncology
4A.03 Predictive Utility of c-reactive Protein (CRP) in Advanced Stage Lung Cancer Treated with AntiProgrammed Cell Death-1 (PD-1) Therapy
D. Pham,1 C. Pinkston,2 M. Oechsli,1 M. Kong,2 J. Rios-Perez,1 G. Kloecker1 1James Graham Brown Cancer Center, University of Louisville, Louisville, KY/US, 2School Of Public Health & Information Sciences, University of Louisville, Louisville, KY/US
Topic: Medical Oncology 1
2
3
4
5
A.R. Naqash, C.R.G. Stroud, C. Cherry, M. Muzaffar, M. Bowling, P.R. Walker6 1Hematology/Oncology, East Carolina University/Vidant Medical Center, Greenville, NC/US, 2Hematology/Oncology, East Carolina University/Vidant Medical Center, Greenville, NC/US, 3Hematology Oncology, East Carolina University Brody School of Medicine, Greenville, NC/US, 4Hematology/Oncology, East Carolina University, Greenville, NC/US, 5Pulmonary Critical Care, East Carolina University Brody School of Medicine, Greenville, NC/US, 6Hematology/Oncology, Brody School of Medicine at East Carolina University, Greenville, NC/US Background: The identification of surrogate biomarkers with a predictive value in patients treated with anti-PD-1 therapy has been a challenge. CRP is an acute phase protein of hepatic origin secreted in response to pro-inflammatory cytokines. Various studies have elucidated an elevated CRP to correlate with inferior outcomes in a multitude of malignancies. We sought to evaluate the predictive utility of CRP in the context of lung cancer treated with anti-PD-1 therapy. Methods: We identified a total of 99 patients with advanced non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) between April 2015 to March 2017 treated with nivolumab after a first line platinum-doublet. Baseline CRP was measured at anti-PD-1 initiation as well as serially with subsequent doses. Cox regression was used to predict associations between survival and candidate inflammatory biomarkers (CRP, albulin) as well as clinical characteristics [age, gender, race, stage, number of metastatic compartments, presence of liver/ brain/bone metastasis, immune-related adverse events (irAEs)]. Overall survival after immunotherapy (OSI) was defined as time from immunotherapy initiation to last follow-up or death. Based on the webbased program designed by Budczies et al, optimal cut-off for CRP in relation to OSl was identified to be 49.1 (approximated to 50 mg/L). Kaplan-Meier method was used to assess survival outcomes. Results: Median age was 65 years with a majority being Caucasian (64.7%) and males (64.6%). NSCLC was predominant (88%), with histology
Background: Every year a significant population exists of those diagnosed with NSCLC who choose not to get treatment upon diagnosis and then “migrate” to additional hospital systems before ultimately getting treatment. Migration to different hospitals may play a role in the decision to treat or not-to-treat, and we aimed to evaluate the potential factors that lead to treatment. Methods: As part of the Kentucky LEADS Collaborative, 27 of 31 Kentucky hospital registries contacted provided NSCLC data of 5,807 out of 10,333 patients from 2012-2014. Variables collected included hospital accreditation status by the Commission on Cancer, patient zip codes, age at diagnosis, stage, overall survival (OS), sex, race, education, income, and insurance status. Treatment included any combination of surgery, radiation, or chemotherapy. Hospital records were matched to Kentucky Cancer Registry records to determine the number of hospitals visited for treatment. Patient treatment and migration patterns were analyzed with a logistic regression model along with additional post-hoc analysis. Results: Treatment, initially or via migration, was more likely when an accredited hospital was visited (84% vs. 59%). Most patients were treated at their initial hospital (73%). However, among the remaining patients, 37% migrated to a different hospital where most then received treatment (92%). Initially treated vs untreated was significantly associated with Stage I-II disease, insurance status, younger age (66.8 vs 70.1 years), and longer OS (506 days vs 306). Migrating to another hospital was associated with Stage IIII disease, younger age (66.4 vs 72.2 years), longer OS (561 vs 157 days), but also notably associated with initial hospitals missing treatment modalities and patients having private insurance. Patients who were treated after migrating were associated with Stage I-II disease, younger age (65.8 vs 72.8 years), and longer OS (595 vs 153 days). Compared to patients who were treated initially, patients treated via migration lived longer (595 vs 506 days) and particularly had longer survival with stage IV disease (414.2 vs 321.6 days). Conclusion: This analysis demonstrates a survival benefit for initially untreated patients who migrate to another hospital. This migration is significantly associated with stage, missing treatment modalities at diagnosis site, and
S1560 insurance status which suggests that patients intending to seek better care will frequently migrate. They are more likely to receive treatment and live longer if they are insured. Considering the current landscape of changing healthcare policy, it is notable that insurance status plays such a significant role in enabling lung cancer patients to find effective treatment.
Journal of Thoracic Oncology
Vol. 12 No. 11S1
4B.03 Prospective Evaluation of Multidisciplinary Lung Cancer Care: Timeliness, Thoroughness, and Patient/ Caregiver Perspectives Topic: Medical Oncology
4B.02 Increasing Survival in Stage IV NSCLC in Academic versus Community Based Centers in the National Cancer Database Topic: Medical Oncology S. Ramalingam,1 M. Dinan,1 J. Crawford2 1Duke University, Durham, NC/US, 2Medical Oncology, Duke University, Durham, NC/US Background: Overall survival of patients in the United States with advanced Non-small cell lung cancer (NSCLC) remains poor. However, improved outcomes are expected from recent advances in management from optimized chemotherapy and biomarker driven personalized therapies. With the increased early use of emerging regimens, we hypothesized that survival differences between academic and community centers are widening over time. Methods: We did a retrospective analysis of patients diagnosed with stage IV NSCLC between 1998 and 2010 within the 2012 National Cancer Database (NCDB). The primary end point was 2-year survival; analysis was limited to patients diagnosed up to 2010 to allow at least 2 years of follow-up. We restricted our analysis to hospitals present in the dataset throughout the study period. Overall survival was compared by academic versus community site of treatment controlling for the year of diagnosis, age, gender, histology, and insurance status. Results: In our multivariable analysis of 193,279 patients with stage IV NSCLC, 2-year survival differentially improved over time between academic and community-based centers (p¼0.0005) at a rate of 0.15% per year with 99.9% confidence interval (0.008%,0.3%). Patients treated in community centers tended to be older (mean 64 versus 62 years) and on average traveled 15 miles less for treatment. These patient differences between community and academic centers were stable over time. Conclusion: We found a widening gap in 2-year survival between academic and community centers in the NCDB. Further research is needed to understand and address widening disparities in outcomes between academic and community centers, and the potential impact of immune approaches on these outcomes.
M. Smeltzer,1 F.E. Rugless,2 H.K. Lee,3 K. Ward,1 N.R. Faris,2 M.A. Ray,1 M. Meadows,1 B. Jiang,2 B. Jackson,2 C. Foust,2 A. Patel,2 N. Boateng,2 S. Kedia,1 K. Roark,2 C. Houston-Harris,2 C. Fehnel,2 R.S. Signore,2 R. Fox,2 E.T. Robbins,2 J. Li,3 R.U. Osarogiagbon2 1 School Of Public Health, University of Memphis, Memphis, TN/US, 2 Baptist Cancer Center, Memphis, TN/US, 3Dept Of Industrial and Systems Engineering, University of Wisconsin-Madison, Madison, WI/US Background: The process of care for Non-small Cell Lung Cancer (NSCLC) patients provides many challenges. A Multidisciplinary (MD) model may improve outcomes, but its value has been difficult to quantify, with most studies being retrospective pre-post analyses. Timeliness of care is one MD advantage that studies consistently demonstrate, but more timely care does not necessarily translate to improved survival. However, timeliness is a patient-centered value, affirmed in our pre-study focus groups. We prospectively evaluated timeliness, staging activities, stage-based treatment selection, and patient/caregiver perspectives in MD vs. usual serial care (SC). Methods: This comparative effectiveness trial evaluated MD clinic vs. SC on multiple patient-centered endpoints with mixed-methods. Timeliness was evaluated by process engineering with bottleneck analyses. The 5-step process of care was: lesion identification, diagnostic biopsy, non-invasive staging, invasive staging, and definitive treatment. Quality of staging endpoints included bimodal staging (CT and PET or invasive biopsy) and trimodal staging (using all 3). Stage based treatments were compared with NCCN recommendations. Patient/caregiver satisfaction with quality of care and timeliness was measured by patient/caregiver surveys at baseline, 3-months, and 6months. Statistical methods: Chi-Square test, Wilcoxon-Mann-Whitney test. Results: The study enrolled 527 patients (178 MD, 349 SC). Average waiting times between steps were 48.8 days (MD) and 43.3 days (SC). Bottlenecks included time from: lesion identification to diagnostic biopsy, invasive or noninvasive staging to definitive treatment. MD patients received higher rates of bimodal staging (89% vs. 76%, p-value<0.001) and tri-modal staging (55% vs. 37%, pvalue<0.001) than SC patients. Ultimately, MD patients received more stage appropriate treatment than SC (79% vs. 68%, p-value<0.01). Overall, patients on MD and SC reported satisfaction with timeliness measures at baseline, 3-, and 6-months, scoring most items similarly high (between 2.7/3.0 and 2.9/3.0; individual question response rates 48-86%). At 6-months, MD patients were more satisfied with overall time from diagnosis to treatment completion (4.9/5.0 vs. 4.6/5.0, pvalue¼0.03). MD patients also reported higher satisfaction with combined quality of care from team members at all 3 time points (6Month: 15.9/16.0 vs. 14.4/16.0, p<0.0001). MD Caregivers were more satisfied with overall quality of care compared with other NSCLC patients at baseline (p<0.01) and 6-months (p<0.01). Conclusion: Despite the additional time to treatment initiation, MD patients received more thorough staging, more stage-appropriate treatment, and were more satisfied with the timeliness of their care. Timeliness is not an absolute quantity, but may be a relative value based on patients’ satisfaction with their level of engagement in the care-delivery process.