Impact of neuroimaging in the pretreatment evaluation of early stage non-small cell lung cancer

Background

There are limited data and conflicting guideline recommendations regarding the role of neuroimaging in the pretreatment evaluation of non-small cell lung cancer (NSCLC).

Methods

We performed a retrospective, pragmatic cohort study of patients with NSCLC diagnosed between January 1 and December 31, 2015. Eligible patients were identified from an institutional tumor registry. We collected all records of pretreatment neuroimaging within 12 weeks of diagnosis, including CT head (CT) and MRI brain (MRI). We abstracted the indication for neuroimaging, presence of central neurologic symptoms and cancer stage (with and without neuroimaging findings) from the tumor registry and the electronic health record.

Results

We identified 216 evaluable patients with newly diagnosed NSCLC. 157 of 216 patients (72.7%) underwent neuroimaging as part of initial staging, and 41 (26%) were found to have brain metastases. Of 43 patients with central neurologic symptoms at the time of neuroimaging, 28 (67%) had brain metastasis. In patients without central neurologic symptoms, brain metastases were discovered in 0 of 33 patients with clinical stage I or II, 4 of 36 (11%) with clinical stage III and 9 of 45 (20%) with clinical stage IV disease.

Conclusions

In patients with early stage NSCLC (i.e. clinical stage I and II) without central neurologic symptoms, brain metastases are unlikely. The continued use of neuroimaging in the pretreatment evaluation of clinical stage I patients without central neurologic symptoms is not needed.

Hospitalization Risk During Chemotherapy for Advanced Cancer: Development and Validation of Risk Stratification Models Using Real-World Data

PURPOSE

Hospitalizations are a common occurrence during chemotherapy for advanced cancer. Validated risk stratification tools could facilitate proactive approaches for reducing hospitalizations by identifying at-risk patients.

PATIENTS AND METHODS

We assembled two retrospective cohorts of patients receiving chemotherapy for advanced nonhematologic cancer; cohorts were drawn from three integrated health plans of the Cancer Research Network. We used these cohorts to develop and validate logistic regression models estimating 30-day hospitalization risk after chemotherapy initiation. The development cohort included patients in two health plans from 2005 to 2013. The validation cohort included patients in a third health plan from 2007 to 2016. Candidate predictor variables were derived from clinical data in institutional data warehouses. Models were validated based on the C-statistic, positive predictive value, and negative predictive value. Positive predictive value and negative predictive value were calculated in reference to a prespecified risk threshold (hospitalization risk ≥ 18.0%).

RESULTS

There were 3,606 patients in the development cohort (median age, 63 years) and 634 evaluable patients in the validation cohort (median age, 64 years). Lung cancer was the most common diagnosis in both cohorts (26% and 31%, respectively). The selected risk stratification model included two variables: albumin and sodium. The model C-statistic in the validation cohort was 0.69 (95% CI, 0.62 to 0.75); 39% of patients were classified as high risk according to the prespecified threshold; 30-day hospitalization risk was 24.2% (95% CI, 19.9% to 32.0%) in the high-risk group and 8.7% (95% CI, 6.1% to 12.0%) in the low-risk group.

CONCLUSION

A model based on data elements routinely collected during cancer treatment can reliably identify patients at high risk for hospitalization after chemotherapy initiation. Additional research is necessary to determine whether this model can be deployed to prevent chemotherapy-related hospitalizations.

Design of eSyM: An ePRO-based symptom management tool fully integrated in the electronic health record (Epic) to foster patient/clinician engagement, sustainability, and clinical impact

e14120

Background: Chemotherapy and surgery can cause distressing symptoms, which can be a burden for health system to address. Programs that directly engage patients, including electronic tracking of patient-reported outcomes (ePROs), can improve symptom control and decrease the need for acute care. Previous ePRO programs have relied on third party vendors with limited EHR integration, constraining their clinical utility and scalability. An integrated solution could offer distinct advantages. Methods: As part of NCI’s Moonshot-funded IMPACT consortium, 6 health systems and Epic built an electronic symptom management program (eSyM) based on the PRO-CTCAE questionnaire that is fully integrated into the EHR. The agile, user-centered design process engaged patients, clinicians, and institutions. The core functional components include: 1) symptom surveys in the postoperative period or between chemotherapy visits, 2) self-management tip sheets, 3) clinician alerts, and 4) dashboards for population management. Critical points of integration with supporting EHR functions and workflow impacts were identified; and major challenges of integration and implementation were described. Results: eSyM, which was implemented at two health systems (Baptist Memorial in Tennessee and Mississippi and West Virginia University Health) in the fall of 2019, required multiple supporting EHR functions: 1) access a secure, HIPPA-compliant patient portal/messaging system (MyChart); 2) record diagnosis, procedure and chemotherapy treatment plan data; 3) identify target populations and track metrics/events; 4) define and execute autonomous logic-based workflow rules; 5) generate reports for clinicians/patients; and 6) documentation. Major challenges included: 1) working within pre-existing EHR system standards and capabilities, which limited the ability to customize interfaces and workflows specifically for the eSyM use case; and 2) adapting to different EHR configurations and polices across multiple health systems. Conclusions: The eSyM build leveraged many existing EHR capabilities and addressed regulatory hurdles; but it required design and workflow compromise. Integration of ePRO-based symptom management programs into the EHR could help overcome barriers, consolidate clinical workflows, and foster scalability/sustainability. Ongoing efforts include launching eSyM at four more sites and evaluating its adoption, usability, and impact on clinical outcomes.

Reliability and correlations among quality measures for lung, breast, and colorectal cancer

2073

Background: Alternative payment models for oncology seek to improve quality and reduce spending. Yet the ability to measure high-quality care across oncology practices remains uncertain. We characterized quality of care for oncology practices using registry and claims-based measures of processes, utilization, end-of-life care, and survival and assessed correlations of practice-level performance across measure type and cancers. Methods: Using SEER-Medicare data, we studied individuals with newly diagnosed lung (N = 95,635), breast (N = 78,736), or colorectal (CRC, N = 51,385) cancers in 2010-2015 treated in oncology practices with ≥20 patients (502, 492, and 347 practices, respectively). We measured receipt of guideline-recommended treatment and surveillance (processes), hospitalizations or emergency department visits during 6-month chemotherapy episodes (utilization), care intensity in the last month of life (EOL), and 12-month survival (lung and CRC only). We calculated summary process, utilization, and EOL measures for each patient (number of measures met divided by the number for which the patient was eligible). We used hierarchical linear models with practice-level random effects to estimate summary measures and survival for each practice. We calculated practice-level reliability (a measurement’s reproducibility) for each measure based on the between-measure variance, within-measure variance, and sample size. Results: Few practices had ≥20 patients eligible for most measures (38%, 37%, and 31% of practices had ≥20 patients for any lung, breast, and CRC measures, respectively). Measure reliability was low. Only 13%, 7%, and 20% of measures for lung, breast, and CRC, respectively, had a median reliability across practices ≥0.7. Among practices with ≥20 patients with summary measures of each type within cancer, correlations across measure types were low (all correlation coefficients (r)≤0.21 except a weak correlation of the CRC process summary measure with 1-year CRC survival, r = 0.38, p < 0.001). Summary process measures were minimally or not correlated across cancer type (lung, breast, CRC; all correlation coefficients ≤0.16). Conclusions: Claims-based measures of care processes, utilization, EOL care, and survival are limited by small numbers of fee-for-service Medicare patients across practices, even after pooling 6 years of data. Measures have poor reliability and are poorly correlated across measure or cancer type. Additional research is needed to identify reliable quality measures for practice-level alternate payment models.

End-of-life quality metrics among medicare decedents at minority-serving cancer centers: A retrospective study

Background

We calculated the performance of National Cancer Institute (NCI)/National Comprehensive Cancer Network (NCCN) cancer centers’ end‐of‐life (EOL) quality metrics among minority and white decedents to explore center‐attributable sources of EOL disparities.

Methods

We conducted a retrospective cohort study of Medicare beneficiaries with poor‐prognosis cancers who died between April 1, 2016 and December 31, 2016 and had any inpatient services in the last 6 months of life. We attributed patients’ EOL treatment to the center at which they received the preponderance of EOL inpatient services and calculated eight risk‐adjusted metrics of EOL quality (hospice admission ≤3 days before death; chemotherapy last 14 days of life; ≥2 emergency department (ED) visits; intensive care unit (ICU) admission; or life‐sustaining treatment last 30 days; hospice referral; palliative care; advance care planning last 6 months). We compared performance between patients across and within centers.

Results

Among 126,434 patients, 10,119 received treatment at one of 54 NCI/NCCN centers. In aggregate, performance was worse among minorities for ED visits (10.3% vs 7.4%, P < .01), ICU admissions (32.9% vs 30.4%, P = .03), no hospice referral (39.5% vs 37.0%, P = .03), and life‐sustaining treatment (19.4% vs 16.2%, P < .01). Despite high within‐center correlation for minority and white metrics (0.61‐0.79; P < .01), five metrics demonstrated worse performance as the concentration of minorities increased: ED visits (P = .03), ICU admission (P < .01), no hospice referral (P < .01), and life‐sustaining treatments (P < .01).

Conclusion

EOL quality metrics vary across NCI/NCCN centers. Within center, care was similar for minority and white patients. Minority‐serving centers had worse performance on many metrics.

Translation of Patient-Reported Outcomes in Oncology Clinical Trials to Everyday Practice

PURPOSE

Clinical trials in oncology evaluating the effects of patient-reported outcomes (PRO) collection have found that monitoring of symptoms with PROs is associated with improved clinical care through reduced acute care utilization and decreased patient symptom burden. This educational review will evaluate strategies for systematic PRO integration into everyday oncology clinical practice.

METHODS

We outline key considerations for using PROs in clinical practice, highlighting evidence from published studies. We also discuss the benefits and challenges of PRO implementation in oncology.

RESULTS

Implementing PRO collection in clinical practice can improve care delivery and facilitate patient-centered clinical research. Considerations for using PROs in clinical practice include choice of instrument, method of delivery, and frequency of query. Challenges with implementing systematic PRO collection include the costs and resources needed for implementation, impact on clinical workflow, and controlling/monitoring physician burnout.

CONCLUSIONS

While challenges exist in terms of financial resources and staff participation/burnout, patient-reported outcomes in clinical practice provide a number of benefits, including symptom monitoring, clinical research, and potential real-time personalized clinical-decision support.

Treatment delays during FOLFOX chemotherapy in patients with colorectal cancer: a multicenter retrospective analysis

Background

FOLFOX (folinic acid, 5-fluorouracil and oxaliplatin) is the most commonly used chemotherapy regimen for the treatment of colorectal cancer. FOLFOX is administered in 14-day cycles, though toxicities frequently lead to unplanned delays. We report the incidence of unplanned delays among patients receiving FOLFOX and describe the reasons for delays.

Methods

We conducted a retrospective analysis of patients receiving FOLFOX chemotherapy for colorectal cancer. Patients were treated at one of two tertiary cancer centers between January 2012 and April 2016. Cycles 2-6 were assessed for delays, and treatments were considered delayed when the interval from prior treatment was >18 days. Reasons for unplanned delays were categorized based on review of clinical records.

Results

We identified 214 patients receiving FOLFOX as standard-of-care therapy. The median age was 59 years, and 55% were female. Of 961 evaluable treatment cycles, 124 (13%) had unplanned delays, and 92 of 214 patients (43%) had one or more unplanned delays in cycles 2-6. Cytopenias (neutropenia and/or thrombocytopenia) were the most common cause of unplanned delays, affecting 34% of patients and accounting for 74 of 124 unplanned delays (60%).

Conclusions

Delays are common during FOLFOX chemotherapy, with 43% of patients having at least one unplanned delay prior to completing cycle 6. Neutropenia and thrombocytopenia were the leading causes of unplanned delays. Our findings justify the development of systematic approaches for preventing unplanned delays, such as standardized laboratory treatment criteria and/or proactive dose adjustment strategies.

Early Findings From the Oncology Care Model Evaluation

PURPOSE:

The Oncology Care Model (OCM) is an alternative payment model administered by the Centers for Medicare & Medicaid Services (CMS) that is structured around 6-month chemotherapy treatment episodes. This report describes the CMS-sponsored OCM evaluation and summarizes early evaluation findings.

METHODS:

The OCM evaluation examines health care spending and use, quality of care, and patient experience during chemotherapy treatment episodes. Because OCM participation is voluntary, the evaluation compares participating physician practices with a propensity-matched group of nonparticipating practices by using a difference-in-differences approach. This report examines 6-month episodes initiated during the first OCM performance period (July 1, 2016, through January 1, 2017).

RESULTS:

During the first OCM performance period, there was no statistically significant impact of OCM on total episode payments. There were small declines in intensive care unit (ICU) admissions (7 per 1,000 episodes) and emergency department visits (15 per 1,000 episodes); there was no statistically significant impact on hospitalizations or 30-day readmissions. Analyses of care quality and end-of-life care showed statistically significant impacts of OCM on the proportion of patients with inpatient hospitalizations in the last 30 days of life (1.5% absolute decrease) and ICU admissions in the last 30 days of life (2.1% decrease). There was no significant OCM impact on measures of hospice use.

CONCLUSION:

Early findings from the OCM evaluation demonstrate modest program-related impacts on some acute care services and no change in total episode payments. Early findings may not reflect practice redesign efforts that were phased in after the beginning of OCM.

Associations between loot box use, problematic gaming and gambling, and gambling-related cognitions

Loot boxes are virtual goods in video games that produce randomly-generated in-game rewards, and have attracted scrutiny because of a resemblance to gambling. This study tests relationships between gaming involvement, engagement with loot boxes, and their associations with disordered gambling and gambling-related cognitions. Online questionnaires were completed by 144 adults via MTurk (Study 1) and 113 undergraduates (Study 2). Gaming and loot box-related variables included estimated time spent gaming and monthly expenditure, the Internet Gaming Disorder Scale (IGDS), and questions that assessed perceptions and behaviours related to loot boxes. Most participants thought loot boxes were a form of gambling (68.1% & 86.2%). A subset of items were condensed into a unidimensional "Risky Loot-box Index" (RLI) via exploratory factor analysis. In Study 1, the RLI showed significant associations with the Problem Gambling Severity Index (r = .491, p < .001) and the Gambling Related Cognitions Scale (r = .518, p < .001). Overall, gambling-related variables predicted 37.1% (p < .001) of the variance in RLI scores. Findings were replicated, though attenuated, in Study 2. These results demonstrate that besides the surface similarity of loot boxes to gambling, loot box engagement is correlated with gambling beliefs and problematic gambling behaviour in adult gamers.

FOLFOX plus ziv-aflibercept or placebo in first-line metastatic esophagogastric adenocarcinoma: A double-blind, randomized, multicenter phase 2 trial

BACKGROUND

Antiangiogenic therapy is a proven therapeutic modality for refractory gastric and gastroesophageal junction adenocarcinoma. This trial assessed whether the addition of a high affinity angiogenesis inhibitor, ziv-aflibercept, could improve the efficacy of first-line mFOLFOX6 (oxaliplatin, leucovorin, and bolus plus infusional 5- fluorouracil) in metastatic esophagogastric adenocarcinoma.

METHODS

Patients with treatment-naive metastatic esophagogastric adenocarcinoma were randomly assigned (in a 2:1 ratio) in a multicenter, placebo-controlled, double-blind trial to receive first-line mFOLFOX6 with or without ziv-aflibercept (4 mg/kg) every 2 weeks. The primary endpoint was 6-month progression-free survival (PFS).

RESULTS

Sixty-four patients were randomized to receive mFOLFOX6 and ziv-aflibercept (43 patients) or mFOLFOX6 and a placebo (21 patients). There was no difference in the PFS, overall survival, or response rate. Patients treated with mFOLFOX6/ziv-aflibercept tended to be more likely to discontinue study treatment for reasons other than progressive disease (P = .06). The relative dose intensity of oxaliplatin and 5-fluorouracil was lower in the mFOLFOX6/ziv-aflibercept arm during the first 12 and 24 weeks of the trial. There were 2 treatment-related deaths due to cerebral hemorrhage and bowel perforation in the mFOLFOX6/ziv-aflibercept cohort.

CONCLUSIONS

Ziv-aflibercept did not increase the anti-tumor activity of first-line mFOLFOX6 in metastatic esophagogastric cancer, potentially because of decreased dose intensity of FOLFOX. Further evaluation of ziv-aflibercept in unselected, chemotherapy-naive patients with metastatic esophagogastric adenocarcinoma is not warranted.

Quality of end-of-life cancer care at minority-serving US cancer centers: A retrospective study of Medicare claims data

6507

Background: Higher EOL treatment intensity is burdensome and has been defined as low-quality care. We explored cancer centers’ EOL quality outcomes among minority and white patients and evaluated whether minority-serving cancer centers had systematically lower EOL quality. Methods: We conducted a retrospective cohort study of Medicare beneficiaries with poor-prognosis cancers who died between April 1, 2016 and December 31 2016. We attributed patients’ EOL treatment to the cancer center where they received the preponderance of inpatient services during the last 6 months of life. We then compared age-sex-comorbidity adjusted center-level measures of EOL care (EOL chemotherapy, emergency department (ED) use, intensive care unit (ICU) admission, hospice use, life-sustaining treatment use, palliative care, and advance care planning) between minority (black, Hispanic, Asian, other) and non-Hispanic white (white) patients within the same center and across centers, grouped by concentration of minorities served (low: < 15%, medium 15-30%, and high > 30%). Results: Among 126,434 patients, 10,006 (21.4% minority) received treatment at one of 53 National Cancer Institute-designated and/or National Comprehensive Cancer Network-affiliated cancer centers. Only 4/8 quality measures had sufficient sample size to calculate a minority-specific rate for ≥10 centers. Those measures showed high within-center correlation for minority and white patients (ICU admission: r = 0.79,p < 0.001; no hospice referral: r = 0.70,p < 0.0001; life sustaining treatment: r = 0.73,p = 0.004; and palliative care: r = 0.78,p < 0.0001), but the mean adjusted rate for minority versus white patients was significantly worse for two measures: no hospice referral (40.2% v. 37.2%; p < 0.02) and life-sustaining treatments (21.8% v. 19.4%; p < 0.02). When grouped by concentration of minorities served (low/medium/high), 5/8 measures showed systematically lower quality as the concentration of minorities increased: more than 1 ED visit (6.0/8.5/7.7%; p = 0.002), ICU admission (29.1/29.7/35.1%; p = 0.0004), no hospice referral (34.3/38.7/36.8% (p = 0.005), and life sustaining treatments (14.8/16.7/17.9%; p = 0.005). Conclusions: There were systematic differences in end-of-life quality measures across US cancer centers. For many measures, quality was lower at centers that served a greater concentration of minorities. However, EOL care quality for minority and white patients was similar for most but not all measures within any given center.

Risk of treatment-related death in carriers of pathogenic DPYD polymorphisms treated with fluoropyrimidine chemotherapy: A systematic review and patient-level analysis

e15132

Background: Polymorphisms of the DPYD gene are present in 3-5% of the population and are associated with increased risk for grade ≥3 toxicity during treatment with fluoropyrimidine (FP) chemotherapy. Fatal toxicities in carriers of DPYD polymorphisms have been described in published reports, however reliable estimates of the risk of treatment-related mortality are lacking. Methods: We conducted a systematic review of the MEDLINE database to identify relevant manuscripts published before January 28, 2018. We searched for published studies of patients receiving standard-dose FP chemotherapy (5-fluorouracil or capecitabine) who had pre-treatment testing for ≥1 of 4 pathogenic DPYD polymorphisms (c.1236G > A/HapB3, c.1679T > G, c.1905+1G > A/*2A, and c.2846A > T) and who were systematically assessed for treatment-related toxicities. In the case of retrospective studies, we required that the cohort be defined by pretreatment characteristics (e.g., patients were not included on the basis of observed toxicity). Two reviewers extracted study- and patient-level data, with discrepancies resolved by consensus. The pooled data were analyzed to estimate the risk of treatment-related mortality among polymorphism carriers. Results: Of the 1290 references screened, 37 publications were included in the final analysis. Patient-level data identified 485 of 14,377 patients (3.4%) with pathogenic DPYD polymorphisms. There were 12 deaths among polymorphism carriers, resulting in a 2.5% risk of treatment-related mortality (95% CI 1.3-4.4%). Only 2 treatment-related deaths were reported in 13,892 patients without identified polymorphisms. Risk of death by genotype is shown in the table; two decedents were compound heterozygotes. Conclusions: Patients with pathogenic DPYD polymorphisms who are treated with standard-dose FP chemotherapy are at significant risk of death and can be prospectively identified through pharmacogenetic testing. [Table: see text]

Classifying Stage IV Lung Cancer From Health Care Claims: A Comparison of Multiple Analytic Approaches

PURPOSE

Cancer stage is a key determinant of outcomes; however, stage is not available in claims-based data sources used for real-world evaluations. We compare multiple methods for classifying lung cancer stage from claims data.

METHODS

Our study used the linked SEER-Medicare data. The patient samples included fee-for-service Medicare beneficiaries diagnosed with lung cancer from 2010 to 2011 (development cohort) and 2012 to 2013 (validation cohort) who received chemotherapy. Classification algorithms considered Medicare Part A and B claims for care in the 3 months before and after chemotherapy initiation. We developed a clinical algorithm to predict stage IV (v I to III) cancer on the basis of treatment patterns (surgery, radiotherapy, chemotherapy). We also considered an ensemble of claims-based machine learning algorithms. Classification methods were trained in the development cohort, and performance was measured in both cohorts. The SEER data were the gold standard for cancer stage.

RESULTS

Development and validation cohorts included 14,760 and 14,620 patients with lung cancer, respectively. Validation analyses assessed clinical, random forest, and simple logistic regression algorithms. The best performing classifier within the development cohort was the random forests, but this performance was not replicated in validation analysis. Logistic regression had stable performance across cohorts. Compared with the clinical algorithm, the 14-variable logistic regression algorithm demonstrated higher accuracy in both the development (77% v 71%) and validation cohorts (77% v 73%), with improved specificity for stage IV disease.

CONCLUSION

Machine learning algorithms have potential to improve lung cancer stage classification but may be prone to overfitting. Use of ensembles, cross-validation, and external validation can aid generalizability. Degradation of accuracy between development and validation cohorts suggests the need for caution in implementing machine learning in research or care delivery.

Comparison of treatment of early-stage breast cancer among Nurses' Health Study participants and other Medicare beneficiaries

PURPOSE

Increasingly epidemiological cohorts are being linked to claims data to provide rich data for healthcare research. These cohorts tend to be different than the general United States (US) population. We will analyze healthcare utilization of Nurses' Health Study (NHS) participants to determine if studies of newly diagnosed incident early-stage breast cancer can be generalized to the broader US Medicare population.

METHODS

Analytic cohorts of fee-for-service NHS-Medicare-linked participants and a 1:13 propensity-matched SEER-Medicare cohort (SEER) with incident breast cancer in the years 2007-2011 were considered. Screening leading to, treatment-related, and general utilization in the year following early-stage breast cancer diagnosis were determined using Medicare claims data.

RESULTS

After propensity matching, NHS and SEER were statistically balanced on all demographics. NHS and SEER had statistically similar rates of treatments including chemotherapy, breast-conserving surgery, mastectomy, and overall radiation use. Rates of general utilization include those related to hospitalizations, total visits, and emergency department visits were also balanced between the two groups. Total spending in the year following diagnosis were statistically equivalent for NHS and SEER ($36,180 vs. $35,399, p = 0.70).

CONCLUSIONS

NHS and the general female population had comparable treatment and utilization patterns following diagnosis of early-stage incident breast cancers with the exception of type of radiation therapy received. This study provides support for the larger value of population-based cohorts in research on healthcare costs and utilization in breast cancer.

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

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.

Is there a common vulnerability in cannabis phenomenology and schizotypy? The role of the N170 ERP

Cannabis use is a known risk factor for the development of psychosis, although the precise nature of this relationship is unclear. The phenomenological experiences associated with cannabis use vary dramatically, and for some resemble certain features of psychosis. We hypothesized that individuals who report particularly unusual experiences associated with cannabis use would demonstrate similar electrophysiological patterns to those who score high on schizotypal personality traits. The Cannabis Experiences Questionnaire (CEQ) and the Schizotypal Personality Questionnaire (SPQ) were used to measure these experiences and traits. A sample of 97 individuals were placed into one of three experimental or two control groups based on their questionnaire scores. These were the "High CEQ", "High SPQ", "High on Both", "Average Users" and "Control" (non-using) groups. Participants completed a visual face perception task. Electroencephalography was used to measure the neural response to the stimuli. The N170 event-related potential (ERP) was used to measure perceptual encoding of the stimulus. The experimental groups elicited significantly reduced N170 ERPs compared to the Control group. The Average User group did not significantly differ from the Control group, and approached significance with the High SPQ group. None of the high scoring groups significantly differed in N170 ERP response from each other. Replicating past research, the CEQ and SPQ scales moderately correlated with each other. The attenuated N170 ERP demonstrated by the high scoring experimental groups may reflect a manifestation of an underlying shared vulnerability.

The Affordable Care Act and Cancer Care Delivery

The Affordable Care Act (ACA) has reformed US health care delivery through insurance coverage expansion, experiments in payment design, and funding for patient-centered clinical and health care delivery research. The impact on cancer care specifically has been far reaching, with new ACA-related programs that encourage coordinated, patient-centered, cost-effective care. Insurance expansions through private exchanges and Medicaid, along with preexisting condition clauses, have helped more than 20 million Americans gain health care coverage. Accountable care organizations, oncology patient-centered medical homes, and the Oncology Care Model-all implemented through the Center for Medicare & Medicaid Innovation-have initiated an accelerating shift toward value-based cancer care. Concurrently, evidence for better cancer outcomes and improved quality of cancer care is starting to accrue in the wake of ACA implementation.

Estimating Costs of Care Attributable to Cancer: Does the Choice of Comparison Group Matter?

OBJECTIVE

To compare alternative strategies for specifying cancer-free control cohorts for estimating cancer-attributable costs of care.

DATA SOURCE, STUDY DESIGN, DATA EXTRACTION

Secondary data analysis of Surveillance, Epidemiology, and End Results data linked to Medicare claims among patients diagnosed with colorectal, lung, breast, and prostate cancers, 2007-2011. We estimated cancer-attributable costs using three alternative reference cohorts: (1) noncancer Medicare patients individually matched by demographic characteristics, (2) noncancer patients individually matched on demographic factors and comorbidity score, (3) cancer patients as their own control, using prediagnosis costs.

PRINCIPAL FINDINGS

Among 44,266 colorectal, 61,584 lung, 55,921 breast, and 67,733 prostate patients, mean total Medicare spending in the first year of diagnosis was $59,496, $54,261, $31,895, and $26,305, respectively. Estimates of cancer-attributable costs ranged from 79 percent to 82 percent of spending for colorectal, 76 percent-79 percent for lung, 65 percent-74 percent for breast, and 60 percent-75 percent for prostate cancers, depending on the reference cohort used. For all cancers, estimates were higher when patients were used as their own control, compared to demographic and comorbidity-matched controls.

CONCLUSIONS

Choice of reference group can have a substantial impact on proportion of total costs attributed to cancer and should be clearly defined in analyses of the costs of cancer care.

Classifying Lung Cancer Severity with Ensemble Machine Learning in Health Care Claims Data

Research in oncology quality of care and health outcomes has been limited by the difficulty of identifying cancer stage in health care claims data. Using linked cancer registry and Medicare claims data, we develop a tool for classifying lung cancer patients receiving chemotherapy into early vs. late stage cancer by (i) deploying ensemble machine learning for prediction, (ii) establishing a set of classification rules for the predicted probabilities, and (iii) considering an augmented set of administrative claims data. We find our ensemble machine learning algorithm with a classification rule defined by the median substantially outperforms an existing clinical decision tree for this problem, yielding full sample performance of 93% sensitivity, 92% specificity, and 93% accuracy. This work has the potential for broad applicability as provider organizations, payers, and policy makers seek to measure quality and outcomes of cancer care and improve on risk adjustment methods.

An administrative stage inference algorithm for use in patients receiving chemotherapy for colorectal cancer

e18121

Background: Claims data are often used for evaluating cancer care outcomes; however, such data lack information about cancer stage. We developed a claims-based stage inference algorithm to classify chemotherapy-treated patients with localized vs. metastatic colorectal cancer (CRC). Methods: We used the SEER-Medicare linked data (2010-‘11) to develop and validate an algorithm to predict cancer stage (localized vs. metastatic) among patients receiving chemotherapy within 6 months of CRC diagnosis. We used claims to identify treatments received (surgery, radiation, and chemotherapy agents) during the 6 months before and after the first dose of chemotherapy. The sample was split 1:1 into development and validation cohorts. After testing in the development cohort, the final algorithm was evaluated in the validation cohort. SEER data served as the gold standard for cancer stage. Results: We identified 25,258 patients with fee-for-service Medicare and a new diagnosis of CRC. 6,907 patients (27%) received chemotherapy for CRC within 6 months of diagnosis. The median age of chemotherapy-treated patients was 73, 49% were female, and 76% were white; 69% had localized cancer (AJCC stage 1-3) and 31% had metastasis at diagnosis (stage 4). Split-sample validation of the final classification algorithm demonstrated sensitivity and specificity of 87% (95% CI 86-89%) and 76% (73-78%) for localized cancer and 73% (70-75%) and 91% (90-92%) for metastatic cancer. The overall accuracy of classification was 83%. Stratified analyses demonstrated preserved algorithm performance across subgroups of age, sex, race, geography, and comorbidity. Misclassification was most common among patients with metastatic disease who were treated with surgery followed by fluoropyrimidine chemotherapy with or without oxaliplatin. 2-year overall survival was 79.8% (stage 1-3) and 35.4% (stage 4) for SEER stage groups, vs. 79.6% and 35.8% for predicted stage groups. Conclusions: A claims-based algorithm can classify extent of disease in chemotherapy-treated CRC patients with an accuracy of 83%. This algorithm will allow more clinically-relevant patient stratification for claims-based evaluations of cancer care outcomes.

Value-Based Medicine and Integration of Tumor Biology

Clinical oncology is in the midst of a genomic revolution, as molecular insights redefine our understanding of cancer biology. Greater awareness of the distinct aberrations that drive carcinogenesis is also contributing to a growing armamentarium of genomically targeted therapies. Although much work remains to better understand how to combine and sequence these therapies, improved outcomes for patients are becoming manifest. As we welcome this genomic revolution in cancer care, oncologists also must grapple with a number of practical problems. Costs of cancer care continue to grow, with targeted therapies responsible for an increasing proportion of spending. Rising costs are bringing the concept of value into sharper focus and challenging the oncology community with implementation of value-based cancer care. This article explores the ways that the genomic revolution is transforming cancer care, describes various frameworks for considering the value of genomically targeted therapies, and outlines key challenges for delivering on the promise of personalized cancer care. It highlights practical solutions for the implementation of value-based care, including investment in biomarker development and clinical trials to improve the efficacy of targeted therapy, the use of evidence-based clinical pathways, team-based care, computerized clinical decision support, and value-based payment approaches.

Emergency Department Management of Patients With Febrile Neutropenia: Guideline Concordant or Overly Aggressive?

OBJECTIVES

The Infectious Diseases Society of America and the American Society of Clinical Oncology recommend risk stratification of patients with febrile neutropenia (FN) and discharge with oral antibiotics for low-risk patients. We studied guideline concordance and clinical outcomes of FN management in our emergency department (ED).

METHODS

Our urban, tertiary care teaching hospital provides all emergency and inpatient services to a large comprehensive cancer center. We performed a structured chart review of all FN patients seen in our ED from January 2010 to December 2014. Using electronic medical records, we identified all visits by patients with fever and an absolute neutrophil count of <1000 cells/mm³ and then included only patients without a clear source of infection. Following national guidelines, we classified patients as low or high risk and assessed guideline concordance in disposition and parenteral versus oral antibiotic therapy by risk category as our main outcome measure.

RESULTS

Of 173 qualifying visits, we classified 44 (25%) as low risk and 129 (75%) as high risk. Management was guideline concordant in 121 (70%, 95% confidence interval [CI] = 63% to 77%). Management was guideline discordant in 43 (98%, 95% CI = 88% to 100%) of low-risk patients versus 9 (7%, 95% CI = 3% to 13%) of high-risk patients (relative risk [RR] = 14, 95% CI = 7.5 to 26). Of 52 guideline-discordant cases, 36 (83%, 95% CI = 72% to 93%) involved low-risk cases with treatment that was more aggressive than recommended.

CONCLUSIONS

Guideline concordance was low among low-risk patients, with management tending to be more aggressive than recommended. Unless data emerge that undermine the guidelines, we believe that many of these hospitalizations and parenteral antibiotic regimens can be avoided, decreasing the risks associated with hospitalization, while improving antibiotic stewardship and patient comfort.