SEER Region Research Articles

Prevalence and timing of chronic obstructive pulmonary disease (COPD) diagnosis among patients newly diagnosed with lung cancer in the US.

118 Background: COPD is one of the most common comorbidities among patients with lung cancer, and an important determinant of their survival. While COPD is frequently under diagnosed in the general population, less is known about the occurrence of COPD underdiagnosis in patients with lung cancer. Objectives: 1) To estimate prevalence and under diagnosis of COPD among patients newly diagnosed with lung cancer. 2) To quantify the association between timing of COPD diagnosis and stage of lung cancer at diagnosis and whether this association is modified by sex, race, and socioeconomic status (SES). Methods: A secondary data analysis was conducted using the SEER-Medicare database. We included Medicare beneficiaries aged 66+years with primary invasive lung cancer diagnosed from 2008 to 2017 who were alive at the time of cancer diagnosis and had continuous Medicare coverage 12 months prior and 3 months after lung cancer diagnosis. Exposure: Comorbid COPD was determined using Medicare claims and classified as early COPD (evidence of COPD related healthcare utilization (HCU) >3 months before lung cancer diagnosis) versus late COPD (evidence of COPD related HCU only during the +/- 3 months from lung cancer diagnosis). Outcome: Early (localized) versus late (regional and distant) stage of lung cancer diagnosis. Statistical analysis: We used generalized linear models to estimate adjusted absolute (prevalence difference PD) and relative (prevalence ratio PR) association. Results: In a population of 185,405 Medicare beneficiaries with lung cancer, 131,230 (71%) had evidence of COPD. Compared to lung cancer patients without COPD, those with COPD were more likely to be younger, male, White, of lower SES, have more comorbidities, higher HCU, and be diagnosed with early stage lung cancer. Patients with early COPD diagnosis constitute 66,867 (51%) of patients with COPD and lung cancer. They were more likely to be older, female, Non-Hispanic White, of lower SES, and to have more comorbidities, acute exacerbation of COPD, and more early-stage lung cancer compared to patients with late COPD diagnosis. We observed a significant positive association between early COPD diagnosis and early-stage lung cancer diagnosis (PR= 1.27, 95% CI= 1.24 to 1.30; PD= 0.062, 95% CI= 0.057 to 0.066) in models adjusted for age, sex, race, SES, HCU, comorbidities, year of cancer diagnosis, SEER region, and histology. In subgroup analyses, this association was stronger for males, Non-Hispanic Black, American Indian Alaskan Native, and patients of lower SES. Conclusions: Approximately seven out of 10 patients with lung cancer have COPD, and late COPD diagnosis is common. Earlier COPD diagnosis is associated with diagnosis of lung cancer at earlier stages. These findings highlight the need for more structured screening of COPD among individuals at high risk for lung cancer.

Racial disparities in treatment initiation between Black and White older adults with multiple myeloma: A SEER-Medicare analysis.

145 Background: We have previously identified racial disparities in multiple myeloma (MM) treatment initiation within 12 months of diagnosis. Black MM patients (60.6%) were significantly less likely to initiate treatment than White MM patients (69.5%). This current study aimed to explore the extent to which Black/White (B/W) disparities in MM treatment initiation were explained by differences in the distribution of demographic and health characteristics. Methods: We identified Black and White adults (66+) diagnosed with MM 2007 to 2017 in linked SEER-Medicare data. Patients were required to have 12 months of continuous Medicare Parts A and B pre- and post-diagnosis and Medicare Part D for 12 months post-diagnosis or until death. Our outcome was MM treatment initiation within 12 months of diagnosis and our independent variable was race (Black or White) as identified in SEER. Covariates thought to confound the relationship between race and MM treatment initiation were added to successive binomial regression models to estimate risk differences and 95% confidence limits (CL). Model 1 included only race. Model 2 added age and sex. Model 3 added Medicare low-income subsidy status and census tract poverty. Model 4 added rurality and SEER region. Model 5 added comorbidity and predicted frailty. Results: Our cohort (n=10,666) consisted of 1,649 Black and 9,017 White older adults with MM. Distributions of all individual, area-level and health characteristics differed significantly by race. In bivariate analysis all characteristics, except for sex and rurality, were significantly associated with treatment initiation. B/W risk differences remained significant across models and ranged from 3.7% (CL=0.9, 6.5) to 10.7% (CL=8.2, 13.2). Covariates explained some, but not all, of the B/W disparities in MM treatment initiation (Table 1). Conclusions: After accounting for differences in demographic and health characteristics, Black MM patients remained significantly less likely to initiate treatment within 12 months. Our study identified several factors that if addressed could attenuate B/W treatment initiation disparities. Further investigation is needed to understand how these factors contribute to treatment disparities, in order to develop health equity interventions to address them. Our results also suggest that there are additional unidentified factors contributing to disparity in MM treatment initiation. These factors must be identified and explored as we work to eliminate B/W disparities in MM treatment.[Table: see text]

Patterns of multispecialty care for low- and intermediate-risk prostate cancer in the use of active surveillance.

Multidisciplinary models of care have been advocated for prostate cancer (PC) to promote shared decision-making and facilitate quality care. Yet, how this model applies to low-risk disease where the preferred management is expectant remains unclear. Accordingly, we examined recent practice patterns in specialty visits for low/intermediate-risk PC and resultant use of active surveillance (AS). Using SEER-Medicare, we ascertained whether patients saw urology and radiation oncology (i.e., multispecialty care) versus urology alone, based on self-designated specialty codes, for newly diagnosed PC from 2010 to 2017. We also examined the association with AS, defined as the absence of treatment within 12 months of diagnosis. Time trends were analyzed using Cochran-Armitage test. Chi-squared and logistic regression analyses were applied to compare sociodemographic and clinicopathologic characteristics between these models of care. The proportion of patients seeing both specialists was 35.5% and 46.5% for low- and intermediate-risk patients respectively. Trend analysis showed a decline in multispecialty care in low-risk patients (44.1% to 25.3% years 2010-2017; P < 0.001). Between 2010 and 2017, the use of AS increased 40.9% to 68.6% (P < 0.001) and 13.1% to 24.6% (P < 0.001) for patients seeing urology and those seeing both specialists respectively. Age, urban residence, higher education, SEER region, co-morbidities, frailty, Gleason score, predicted receipt of multispecialty care (all P < 0.02). Uptake of AS among men with low-risk PC has occurred primarily under the purview of urologists. While selection is certainly at play, these data suggest that multispecialty care may not be required to promote the utilization of AS for men with low-risk PC.

Child Anxiety, Depression, and Post-traumatic Stress Disorder Following Orthopedic Trauma.

Many children and adolescents are exposed to different types of trauma, e.g., abuse or various disasters. Trauma can cause severe and long-term impairment and consequences, the most studied of which are post-traumatic stress disorder (PTSD) and PTSD symptoms (PTSS). PTSD is highly prevalent in clinical practice (with a frequency of about 7%) and is a debilitating consequence of trauma. The current study aimed to assess childhood injuries and their associated anxiety, depression, and post-traumatic stress disorder following orthopedic trauma. A descriptive cross-sectional study was conducted, including all pediatric patients with trauma at Abha Maternity and Children Hospital in the Seer region of Saudi Arabia, as well as pediatric patients with trauma at Abha Maternity and Children Hospital during the period from January 1, 2021 to December 31, 2022. Data were collected from the children's caregivers using a direct interview questionnaire to assess the children's personal data, depression, anxiety, and post-traumatic stress disorder. Children's trauma-related data were extracted from their medical records using a pre-structured data extraction sheet. A total of 100 children with trauma were included. Children ranged in age from eight to 12 years, with a mean age of 7.3 ± 3.4 years. In all, 67 (67.0%) children were males, and only 6 (6.0%) had chronic health problems. The vast majority of the children with trauma had a low-severity experience of depression and anxiety following trauma (97.1% for each), and only one child had a high-severity experience of depression and anxiety. In all, 5 (4.9%) children with trauma experienced clinically significant PTSD, and the vast majority of them showed a low likelihood of the disorder. Multiple fractures and undergoing surgery were significant predictors of developing PTSD (P < 0.05). In conclusion, the current study revealed that bone trauma was frequent among children, mainly due to playing accidents. Also, a low prevalence of post-traumatic stress disorders and their mental consequences was estimated.

Racial differences in the utilization of MRI-guided biopsy before prostate cancer diagnosis.

5093 Background: Most low-risk prostate cancer (PCa) patients undergo active surveillance (AS). Compared with Whites, Blacks are more likely to opt for AS and suffer a higher risk of cancer progression while on AS. A primary concern of AS is the likelihood of missing clinically significant cancer. Transrectal ultrasound-guided biopsy is the most common method for diagnosing and grading PCa; however, this method often misses anterior PCa. Magnetic resonance imaging (MRI) allows clinicians to identify better and detect PCa; however, current guidelines on MRI-guided biopsy (MRI-GB) are inconsistent. This population-based study aimed to examine the utilization of MRI-GB 6 months before PCA diagnosis over time by race in the US. Methods: We used SEER-Medicare data to identify men diagnosed with PCa between January 1, 2012, to December 31, 2017, and excluded patients missing the date of diagnosis, being diagnosed at death, enrolled in HMO, or having no continuous A and B coverage six months before diagnosis. We used the CPT codes to identify prostate biopsy and MRI use. Multivariable logistic regression was used to evaluate racial disparity in MRI use six months before PCa. The model included covariates age, race, SEER region, marital status, state buy-in for Medicare (a poverty indicator), education, income, Charlson comorbidity index, and procedure (MRI) year. Results: Among 38,612 eligible men, MRI use six months before PCa diagnosis increased from 2012 to 2017 (3.2% to 24.3% among Whites, 1.8% to 14.2% among Blacks). Compared with Whites, Blacks were 38% less likely to receive pre-diagnosis MRI (OR=0.62, 95% CI 0.52-0.74). Besides race, having state buy-in for Medicare (OR=0.44; 95% CI 0.30-0.65) and geographic regions (Central vs. West OR=0.40; 95% CI 0.34-0.45) were strong predictors of lower utilization of MRI. Conclusions: The gap in the utilization of MRI may exacerbate racial disparities in PCa. Future studies are needed to identify potential barriers to adopting MRI for PCa workups.

Health-related quality of life among metastatic prostate cancer survivors: Comparisons to survivors originally diagnosed with localized disease and men without a cancer history in the United States.

205 Background: Maximizing health-related quality of life (HRQOL) is an important therapeutic goal for metastatic prostate cancer patients. Moreover, HRQOL measures can inform clinicians about treatment sequelae and aide decision-making. To date, few studies have comprehensively compared HRQOL between metastatic prostate cancer survivors and survivors with localized diseases and men without a cancer history. Methods: The SEER-Medicare Health Outcomes Survey (MHOS) data linkage was used to identify prostate cancer survivors and individuals without a cancer history enrolled in Medicare Advantage plans. Prostate cancer survivors (diagnosed 1988-2017) who completed the MHOS baseline survey (1998-2019) after diagnosis but &lt; 10 years post diagnosis were included, as were men without a cancer history who completed the MHOS in the same years. Survey records were clustered at person-level. 27,829 prostate cancer survivors with 42,277 survey records (metastatic n = 752 with 1040 records) and 784,305 Medicare Advantage enrollees in SEER regions without a cancer history with 1.15 million survey records were identified. Multilevel linear regressions were used to compare HRQOL outcomes, i.e. The Veterans RAND 12 Item Health Survey (VR-12 scores), between metastatic survivors and other survivors, and men without a cancer history. The VR-12 includes 8 health domains: general health, physical functioning, role limitations due to physical and emotional problems, bodily pain, energy-fatigue, social functioning and mental health. All analyses adjusted for age at survey, stage, comorbid conditions, body mass index, race/ethnicity, marital status, socioeconomic status, SEER region, and survey era. Results: Compared to men without a cancer history, prostate cancer survivors were older, had fewer comorbid conditions, were more likely to racial/ethnic minorities, married, and higher socioeconomic status. Compared to men without a cancer history, metastatic prostate cancer survivors were most likely to report worse general health (T-score difference: -6.26, 95% confidence internal [95CI]: -7.14 to -5.38; p &lt;.001), and physical (T-score difference: -4.33, 95CI: -5.18 to -3.48; p &lt;.001) and mental component summary (T-score difference: -2.64, 95CI: -3.40 to -1.88; p &lt;.001), followed by survivors with early stage diseases in adjusted analyses. Results were similar for other HRQOL VR-12 measures. Conclusions: Metastatic prostate cancer survivors experience significantly worse HRQOL than men diagnosed with early-stage disease and men without a cancer history across all domains. Our findings suggest a need for better symptom management and palliative care for men diagnosed with metastatic prostate cancer.

The effect of race/ethnicity on cancer-specific mortality after salvage radical prostatectomy.

BackgroundTo test the effect of race/ethnicity on cancer-specific mortality (CSM) after salvage radical prostatectomy (SRP).Material and methodsWe relied on the Surveillance, Epidemiology and End Results database (SEER, 2004–2016) to identify SRP patients of all race/ethnicity background. Univariate and multivariate Cox regression models addressed CSM according to race/ethnicity.ResultsOf 426 assessable SRP patients, Caucasians accounted for 299 (69.9%) vs. 68 (15.9%) African-Americans vs. 39 (9.1%) Hispanics vs. 20 (4.7%) Asians. At diagnosis, African-Americans (64 years) were younger than Caucasians (66 years), but not younger than Hispanics (66 years) and Asians (67 years). PSA at diagnosis was significantly higher in African-Americans (13.2 ng/ml), Hispanics (13.0 ng/ml), and Asians (12.2 ng/ml) than in Caucasians (7.8 ng/ml, p = 0.01). Moreover, the distribution of African-Americans (10.3%–36.6%) and Hispanics (0%–15.8%) varied according to SEER region. The 10-year CSM was 46.5% in African-Americans vs. 22.4% in Caucasians vs. 15.4% in Hispanics vs. 15.0% in Asians. After multivariate adjustment (for age, clinical T stage, lymph node dissection status), African-American race/ethnicity was an independent predictor of higher CSM (HR: 2.2, p < 0.01), but not Hispanic or Asian race/ethnicity. The independent effect of African-American race/ethnicity did not persist after further adjustment for PSA.ConclusionAfrican-Americans treated with SRP are at higher risk of CSM than other racial/ethnic groups and also exhibited the highest baseline PSA. The independent effect of African-American race/ethnicity on higher CSM no longer applies after PSA adjustment since higher PSA represents a distinguishing feature in African-American patients.

Sa1240 PREDICTORS FOR LYMPH NODE METASTASIS OF T1B ESOPHAGEAL CANCER AND SURVIVAL FOLLOWING ENDOSCOPIC THERAPY WITH/WITHOUT ADJUVANT THERAPY AND SURGERY: AN ANALYSIS OF SURVEILLANCE, EPIDEMIOLOGY, AND END RESULTS (SEER) DATABASE

Endoscopic therapy (ET) has become a treatment of choice for high grade dysplasia and intramucosal (T1a) esophageal cancer (EC). For T1b EC (tumor invading submucosa), esophagectomy remains the mainstay treatment because previous studies, using single-centered data, reported higher risk of lymph node metastasis (LNM). The use of ET in low risk T1b EC and ET followed by adjuvant chemotherapy ± radiation (CRT) for patients who decline or are unfit for surgery have been reported. However, outcomes data have been limited to a few small studies from tertiary centers. 1) to determine the prevalence and risk factors of LNM of T1b EC using a large national cancer database and 2) evaluate the survival outcomes of T1bN0M0 EC after ET alone, ET+CRT compared to radical surgery. We used the SEER 18 database to identify patients with T1b histologically confirmed EC diagnosed between 1998-2016. Prevalence of LNM at initial diagnosis and its risk factors were analyzed. Cancer-specific survival (CSS) and overall survival (OS) were calculated using Kaplan-Meier’s (KM) estimate. The multivariate analysis was performed with the Cox-regression method after controlling for other relevant covariates. A total of 1374 patients with T1b EC were identified (mean age: 65; 82% men; 89% white; 77% adenocarcinoma[AC]; 22% squamous cell cancer[SCC]) (Table 1).Overall, 283 patients(20.6%) had LNM at the time of diagnosis with prevalence of 20% in AC and 22.8% in SCC(p=0.279). In multivariate analysis, age≦65, diagnosis in 2010-2016(vs older years), tumor size ≧ 2 cm, and poorly differentiation were associated with LNM. Survival of patients with T1b with LNM was significantly lower than those with LNM(-) [3-yr OS: 63.7% and 47.8%, p<0.001; 3-yr CSS: 72.1% and 54.1%, p<0.001]. Subgroup of patients with T1bN0M0 EC who received ET only(n=96), ET+CRT(n=22) and surgery(n=788) were further analyzed. The median follow-up time was 39 months(IQR 14-89). 3-yr OS and CSS were: 65%/81% in ET, 30%/88% in ET+CRT and 73%/79% in surgery. KM curves are shown in Figure 1. In a multivariate proportional hazards model; adjusted for age, gender, race, histologic type, year of diagnosis, site of cancer, SEER region, ET alone and surgery resulted in statistically equivalent in overall mortality[Hazard ratio(HR) 0.93, 95%CI 0.68-1.43, p=0.93] and esophageal cancer mortality[HR 1.47, 95%CI 0.86-2.51, p=0.16]. ET+CRT, as compared to ET alone, was associated with increased overall mortality[HR 2.17, 95%CI 1.03-4.57, p=0.04], with no significant difference in esophageal cancer survival. In this population-based study, the prevalence of LNM in T1b EC is 20%. In patients with T1bN0M0, overall and cancer specific survival of those treated with ET appears to be similar to those treated with surgery. Addition of CRT following ET negatively impacts survival outcomes.Figure 1Kaplan–Meier graphs showing the survival in patients with T1bN0M0 esophageal cancer treated with endoscopic treatment +/- chemoradiation or surgeryView Large Image Figure ViewerDownload Hi-res image Download (PPT)

HSR19-111: Disparities in Guideline-Concordant Care Among HER2 Positive Metastatic Breast Cancer Patients

Background: We examined guideline-concordant care among women with HER2+ MBC and determined the magnitude of differences in guideline-concordant care between those with positive and negative hormone receptor (HR) status by utilizing a non-linear decomposition technique. Methods: We conducted a retrospective observational cohort study using the Surveillance, Epidemiology, End Results-Medicare linked database. The study cohort consisted of women age &gt;66 years diagnosed with HER2+ MBC in 2010–2013 (N=241). Guideline-concordant initial care within 6 months of cancer diagnosis was defined as per NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). A multivariate logistic regression was performed to identify the significant predictors of guideline-concordant care. A post-regression non-linear decomposition was conducted to examine the magnitude of disparities in guideline concordant care by women’s HR status. Results: 76.8% of the study cohort received guideline-concordant care, while 23.2% did not. As compared to those who did not receive guideline-concordant care, women who received guideline-concordant care were significantly more likely to have positive HR status (adjusted odds ratio (AOR)=2.11; P=.04), had good performance status (AOR=3.46; P=.0008), and had a higher number of oncology visits (AOR=8.05; P&lt;.0001). With 1 year increase in age at cancer diagnosis, there was 5% lesser likelihood of receiving guideline-concordant care (AOR=0.95; P=.04). From the decomposition analysis, 19.0% of the disparity in guideline-concordant care between women with positive and negative HR status was explained by differences in their characteristics. Enabling characteristics (marital status, census-level income, and education) explained the highest (22.8%) proportion of the disparity, followed by external environmental factors (location of residence, SEER region, hospitals offering oncology services) at 5.3%, and need-related factors (tumor grade, comorbidity, performance status, number of metastases) at 3.2%. Conclusion: Almost one quarter of the study cohort did not receive guideline-concordant care. There are opportunities to improve cancer care for women with negative HR status who have lower socioeconomic status. The high unexplained portion of differences in guideline-concordant care (81.0%) can be due to patient preferences for treatment, propensity to seek care, and organizational and physician-level factors not captured in the database.

Comparative effectiveness of primary radiotherapy versus surgery in oropharyngeal squamous cell carcinoma in elderly patients: A SEER-Medicare analysis.

e18124 Background: Primary radiotherapy (RT) and primary surgery (PS) are considered equally viable local therapy modalities for oropharyngeal squamous cell carcinoma (OPSCC). The relative efficacy and toxicity of these competing therapies are often debated, and treatment choice is based on a paucity of comparative data. Methods: Eligible individuals were elderly patients in the SEER-Medicare registry diagnosed with OPSCC between 2000 and 2011. Patients were categorized as receiving either primary RT +/- chemotherapy, or PS +/- adjuvant RT or chemoradiotherapy (CRT). Overall survival (OS) was analyzed using Cox multivariable analysis (MVA). Risks of gastrostomy dependence (GD), esophageal stricture (ES), and osteoradionecrosis (ORN) at 1 year were determined through claims and analyzed using logistic regression. Results: A total of 3,804 patients (70% RT, 30% PS) were included in this cohort. At a median follow-up of 24.2 months, median and 2-year OS were 37.9 months and 61.1%, respectively. Patients treated with RT and PS experienced 2-year OS and median survival outcomes of 58.5%/34.9 months and 66.4%/48.0 months, respectively (p &lt; 0.001). In early-stage OPSCC, MVA found older age, unmarried status, increasing comorbidity, SEER region, tumor site, and no prior PET associated with inferior OS; treatment was not associated with OS. For locally advanced OPSCC, increasing age, unmarried status, increasing comorbidity, lower income, tumor site, higher stage, no prior PET, and RT alone (HR = 1.24 for RT alone vs. PS, p = 0.0020, whereas HR 1.08 for CRT vs. PS, p = 0.1955) were associated with inferior OS. Multivariable predictors of ES (9.5% of cohort) were SEER region, grade, and PS treatment (OR = 1.03 and 0.70 for RT and CRT vs. PS, p = 0.015). Independent predictors of GD at 6 months (21.5%) included married status, smoking, tongue site, advanced stage, and CRT treatment (OR = 1.13 and 2.30 for RT and CRT vs PS, p &lt; 0.001). The 1 year risk of ORN was 3.8% and not associated with treatment. Conclusions: Without a clearly dominant modality in elderly patients, local therapy decisions for OPSCC must be individualized to each patient’s disease and functional status.

Association between lymph node examination and nodal staging for pancreatic adenocarcinoma: A population-based analysis.

e15772 Background: The minimum number of examined lymph nodes (ELNs) for accurate nodal staging remains controversial and has not been recommended by the NCCN guidelines for pancreatic adenocarcinoma (PADC). We aimed to evaluate the time trends in the ELN count and the proportion of node-positive PADC (N+%), as well as determine the minimal threshold for the ELN count. Methods: We analyzed 9258 patients with resectable PADC between 2000 and 2012 in the SEER database (18 registries). Changes in the ELN count and N+% over time were evaluated using the Spearman correlation analysis and the Cochran-Armitage test, respectively. Multivariable logistic and Cox models were used to evaluate the association between the ELN count and N+% and overall survival after adjusted for clinicopathologic features (i.e., race/ethnicity, sex, age, marital status, SEER region, year of diagnosis, tumor grade, tumor site, 8th AJCC T stage, and surgical type). Restricted cubic spline modeling was used to handle possible nonlinear effects of the ELN count as a continuous covariate in the multivariable models. Structural breakpoints, which were considered the threshold of clinical impact, were determined by the Chow test. Results: The median ELN count increased from 9 (interquartile range: 5−15) in 2000 to 16 (11−23) in 2012 ( Ptrend &lt; 0.001), and the N+% increased from 61.1% to 67.2% ( Ptrend &lt; 0.001). An increased ELN count was associated with a nonlinear increase in N+% ( P &lt; 0.001), and the Chow test revealed a threshold ELN count of 8 (ORadjusted for ≥ 8 vs. &lt; 8 ELNs, 2.42 [95% CI, 2.19−2.68]). The 8-ELN threshold was confirmed in patients with declared N0 disease: A larger number of ELNs was correlated with a nonlinear increase in overall survival among these patients ( P &lt; 0.001), with the threshold ELN count also being 8 (HRadjustedfor ≥ 8 vs. &lt; 8 ELNs, 0.79 [95% CI, 0.75−0.87]). Moreover, the 8-ELN threshold remained valid regardless of T stage and year of diagnosis. Conclusions: Lymph node examination has increased markedly over time and was associated with an overall shift toward higher-staged PADC. We recommend 8 ELNs as the cut point for improvement of prognostic accuracy and quality measurement of lymph node examination.

Determinants of spending for metastatic breast, lung, and colorectal cancer in SEER-Medicare.

6621 Background: A substantial proportion of cancer spending is directed towards patients with metastatic disease. Past efforts to characterize spending for metastatic cancer have been limited, because they have not included patients with recurrent disease or analyzed spending across the entire episode of care. Spending for stage IV and recurrent metastatic cancer patients may differ. Methods: Using SEER-Medicare data from 2008-13, we identified breast (BC), colorectal (CRC), and lung (LC) cancer patients who were continuously enrolled in parts A, B and D, and had either stage IV or recurrent disease (i.e., return of cancer after resection of stage I-III disease). Mean total Medicare spending/patient per month and per year (2012$US) were estimated from 12 months prior to 12 months after diagnosis, and described for relevant patient sub-groups. Results: In a cohort of 27,847 patients, total spending for stage IV vs. recurrent cancer was 61-73% lower in the year before diagnosis ($11,339 vs. $28,796 for BC; $13,359 and $49,804 for CRC; $15,118 and $49,555 for LC), and 28-88% higher in the year after diagnosis ($68,787 and $42,091 for BC; $111,304 and $58,657 for CRC; $92,181 and $72,354 for LC). When considering the 2 year-period spanning the diagnosis, spending was similar (≤14%) between groups. The primary drivers of spending differences between patients with stage IV and recurrent disease were cancer type and time from diagnosis (Table). Younger age, higher comorbidity, and SEER region were also drivers of higher spending, especially after diagnosis. Conclusions: Spending patterns differ for patients with stage IV vs. recurrent cancer, suggesting different patterns of care that warrant further investigation. Spending differences after diagnosis were driven largely by part B spending, which was due in part to differential chemotherapy use. [Table: see text]

Abstract P3-12-02: Patient prognostic score and survival benefit offered by radiotherapy for ductal carcinoma in situ

Abstract Background: In general, radiotherapy (RT) follows breast-conserving surgery (BCS) and remains the standard of care for the surgical management of both invasive carcinoma and ductal carcinoma in situ (DCIS). Although it is associated with better local control, the magnitude of survival benefit conferred by RT for DCIS has not yet been established. We sought to evaluate whether a survival benefit exists with the addition of RT for patients with DCIS and to validate a patient prognostic score to predict survival benefit. Methods: We performed a retrospective longitudinal cohort study by using the Surveillance Epidemiology and End Results database (SEER 17). Between 1988-2007, we identified 32,144 eligible patients who underwent BCS for DCIS. Using age, year of diagnosis, race, tumor size, hormone receptor status, tumor grade, marital status and SEER region, we calculated propensity score weights to balance clinicopathologic factors between patients receiving only surgery and those receiving surgery and RT. This cohort was divided into seven groups according to the previously validated patient prognostic score proposed by Smith et al. Breast cancer mortality (BCM) was assessed using a log-rank test and a multivariable Cox proportional hazards model. Results: Of 32,144 cases of DCIS, 20,329 cases (63%) were treated with RT (+RT group) and 11,815 cases (37%) were treated with surgery alone (-RT group). There were 304 breast cancer-specific deaths observed over the follow-up period (median 96 months). The weighted cumulative incidence of BCM at ten-years was 1.8% for the +RT group compared to 2.1% for the -RT group (p= 0.003). The effect of RT on survival differed by nuclear grade (p= 0.007), age (p= 0.004), and tumor size (p=0.02). We found that the survival benefit for the +RT group was significantly greater than for the –RT group in subgroups of patients with higher nuclear grade, younger age, and larger tumor size, whereas a statistical reduction of BCM with RT was not observed among patients without these prognostic factors. Moreover, the magnitude of survival benefit was significantly correlated with the patient prognostic score [p&amp;lt;0.0001, Table]. Conclusion: In this population-based cohort study, the patient prognostic score for DCIS accurately estimated the magnitude of survival benefit offered by radiotherapy after BCS, suggesting that decisions for RT could be tailored based on prognostic score and patient preference. Limitations of this study include unmeasured confounders such as a lack of information about patients' comorbidities, margin status and endocrine therapy, and further external validation is needed to confirm our results. Patient Prognostic Score and Hazard Ratio (HR) Comparing Mortality between Radiotherapy Group and non-Radiotherapy GroupPatient Prognostic ScoreNumber of patients in -RT groupNumber of patients in +RT groupWeighted HR of BCM95% CIWeighted HR of OM95% CI078213881.20.67 - 2.10.910.76 - 1.11267744801.00.70 - 1.50.880.78 - 0.992410570800.690.51 - 0.940.710.63 - 0.793304854170.730.48 - 1.10.680.58 - 0.81496517010.310.16 - 0.580.420.30 - 0.5852232480.290.09 - 0.910.430.21 - 0.9161515N.A. N.A. Abbreviation: RT, Radiotherapy; BCM, Breast Cancer Mortality; OM, Overall Mortality: N.A., not available Citation Format: Sagara Y, Freedman RA, Vaz-Luis I, Mallory MA, Wong S, Aydogan F, DeSantis S, Barry WT, Golshan M. Patient prognostic score and survival benefit offered by radiotherapy for ductal carcinoma in situ. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-12-02.

Predictors of radiation therapy (RT) use among Medicare patients with metastatic non-small cell lung cancer (NSCLC).

124 Background: Palliative RT can improve quality of life, but also incurs time and financial costs. We evaluated factors associated with use and intensity of palliative RT for metastatic NSCLC. Methods: Using SEER-Medicare data, we identified patients diagnosed with pathologically confirmed stage IV NSCLC from 2004-2011. Receipt of RT and total fractions in the first RT course was collected. Univariable and multivariable logistic regression were used to identify factors associated with RT receipt and higher intensity RT ( &gt; 10 fractions). Models were adjusted for clustering at the county level. Results: Among 53,371 patients with stage IV NSCLC, 39% (20,937) received palliative RT. Among patients receiving RT, 48% (10,048) received &gt; 10 fractions and only 3.5% (724) received single fraction RT. On multivariable analysis, factors associated with greater RT use included younger age group (overall p &lt; .01), lower Charlson comorbidity score (overall p &lt; .01), female gender (OR 1.1, p &lt; .01), married status (OR 1.1, p &lt; .01), South (OR 1.0) and Midwest (OR 1.1) SEER region vs. Northeast (ref) or West (OR 0.9) (overall p &lt; .01), higher radiation oncologist density tertile (OR 1.1, overall p = .02), hospice use (OR 1.47, p &lt; .01), and chemotherapy use (OR 3.61, p &lt; .01). Among patients treated with RT, predictors for &gt; 10 fractions were older age group (overall p = .03), no hospice use (OR 1.1, p &lt; .01), chemotherapy use (OR 1.9, p &lt; .01) and freestanding vs. hospital-based RT facility (OR 1.66, p &lt; .01). Conclusions: A combination of patient, treatment, and health service factors influenced whether a patient received palliative RT. In contrast, most patient factors (except age) did not influence the decision to treat with higher intensity radiation. The receipt of chemotherapy and treatment at a freestanding RT facility were highly predictive for extended RT fractionation, which may have been partially driven by higher reimbursement rates to freestanding facilities during this period.

MP9-01 COMPARATIVE EFFECTIVENESS OF AGGRESSIVE VS NON-AGGRESSIVE TREATMENT FOR HIGH-GRADE PROSTATE CANCER ACROSS DIFFERENT COMORBIDITY LEVELS

You have accessJournal of UrologyGeneral & Epidemiological Trends & Socioeconomics: Evidence-based Medicine & Outcomes II1 Apr 2015MP9-01 COMPARATIVE EFFECTIVENESS OF AGGRESSIVE VS NON-AGGRESSIVE TREATMENT FOR HIGH-GRADE PROSTATE CANCER ACROSS DIFFERENT COMORBIDITY LEVELS Timothy Daskivich, Julie Lai, Andrew Dick, Claude Setodji, Janet Hanley, Mark Litwin, and Christopher Saigal Timothy DaskivichTimothy Daskivich More articles by this author , Julie LaiJulie Lai More articles by this author , Andrew DickAndrew Dick More articles by this author , Claude SetodjiClaude Setodji More articles by this author , Janet HanleyJanet Hanley More articles by this author , Mark LitwinMark Litwin More articles by this author , and Christopher SaigalChristopher Saigal More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2015.02.361AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Older men with multiple comorbidities have high rates of death from causes other than prostate cancer within ten years of diagnosis, but they may benefit from aggressive therapy for high-grade disease. We sought to compare effectiveness of aggressive (surgery, radiation therapy, brachytherapy) vs. non-aggressive (watchful waiting, active surveillance) treatment in reducing cancer-specific mortality among older men with high-grade prostate cancer across differing levels of comorbidity. METHODS We sampled 44,521 men from the SEER-Medicare database, aged 66 and older, with clinically localized, poorly-differentiated prostate cancer diagnosed between 1991 and 2007. We used propensity-adjusted competing risks regression analysis to compare the risk of cancer-specific mortality between those treated aggressively and non-aggressively among comorbidity subgroups. We calculated 5-, 10-, and 15-year cancer mortality among those treated aggressively and non-aggressively for each comorbidity subgroup, determined 10-year absolute risk reduction in cancer mortality, and calculated the number needed to treat to save one life at ten years. Models were adjusted for age at diagnosis, race, stage, SEER region, socioeconomic status, marital status, and propensity for receiving aggressive treatment. RESULTS In propensity-adjusted competing risks regression analysis, aggressive treatment was associated with a significantly lower risk of cancer-specific mortality in men with high-grade tumors for those with Charlson scores of 0 (HR 0.43, 95% CI 0.39-0.47), 1 (HR 0.48, 95% CI 0.40-0.58), and 2 (HR 0.46, 95% CI 0.34-0.62) but not in men with Charlson scores of 3+ (HR 0.68, 95% CI 0.44-1.07). Absolute reduction in 10-year cancer specific mortality between those treated aggressively and non-aggressively was 11.3%, 7.9%, 8.6%, and 2.8% for men with Charlson scores of 0, 1, 2, and 3+, respectively. The numbers needed to treat to prevent one cancer death at ten years were 9, 13, 12, and 36 men for those with Charlson scores of 0, 1, 2, and 3+. CONCLUSIONS Older men with Charlson scores ≤2 should consider aggressive treatment of high-grade prostate cancer due to the substantial, short-term cancer-specific survival benefits of aggressive treatment and the relatively low number needed to treat to prevent one cancer death at ten years. Those with Charlson scores of 3 or greater may wish to pursue conservative management, given minimal benefit from aggressive treatment. © 2015 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 193Issue 4SApril 2015Page: e102 Advertisement Copyright & Permissions© 2015 by American Urological Association Education and Research, Inc.MetricsAuthor Information Timothy Daskivich More articles by this author Julie Lai More articles by this author Andrew Dick More articles by this author Claude Setodji More articles by this author Janet Hanley More articles by this author Mark Litwin More articles by this author Christopher Saigal More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

AT-08 * RADIOTHERAPY WITH VS WITHOUT TEMOZOLOMIDE IN OLDER PATIENTS WITH GLIOBLASTOMA

BACKGROUND: A randomized trial of glioblastoma patients aged ≤70 years demonstrated that adding temozolomide (TMZ) to radiotherapy (RT) improved overall survival (OS). We examined whether adding TMZ to RT was associated with improved OS in older glioblastoma patients. METHODS: We performed a retrospective cohort study of 1,718 Medicare beneficiaries aged ≥65 years with glioblastoma diagnosed in a SEER region who received ≥10 daily fractions of RT between 2005-2009, or between 1995-1999 before TMZ was available. Patients were categorized into 3 cohorts based on diagnosis year and treatment received within 3 months of diagnosis: 1) 2005-2009 and TMZ/RT, 2) 2005-2009 and RT only, or 3) 1995-1999 and RT only. Associations between treatment group and OS were compared using Cox proportional hazards models and propensity score analyses, controlling for baseline patient characteristics. RESULTS: Median survival estimates were 9.4 (IQR, 5.2-17.0) months for TMZ/RT, 7.9 (IQR, 4.7-15.0) months for RT in 2005-2009, and 7.7 (IQR, 4.9-11.8) months for RT in 1995-1999. Two-year survival probabilities were 12.4% for TMZ/RT, 7.6% for RT in 2005-2009, and 3.3% for RT in 1995-1999. Propensity score-adjusted models showed decreased mortality risk for TMZ/RT compared to RT in 2005-2009 (AHR, 0.88; 95% CI, 0.79-0.98) and to RT in 1995-1999 (AHR, 0.70; 95% CI, 95% CI, 0.56-0.87). Subgroup and sensitivity analyses showed no significant advantage to TMZ/RT compared to RT in 2005-2009 among patients receiving long-course RT (≥27 fractions; AHR, 0.89; 95% CI, 0.79-1.01). CONCLUSIONS: Concurrent TMZ was associated with a small survival benefit among older glioblastoma patients receiving RT, but no significant survival benefit was observed among patients who received standard long-course RT.

Surveillance imaging following colon cancer resection: Universal or risk-based?

307 Background: Current colon cancer surveillance guidelines recommend annual imaging for all patients following curative resection. The primary aim of this study is to evaluate variation in surveillance practices following colon cancer resection. Methods: A retrospective cohort study of patients ≥ 66 years old with stage I-III colon cancer that underwent surgical resection in the Surveillance, Epidemiology, and End Results-Medicare linked database (July 2001 through December 2007) with at least 3 years of follow-up was performed. Medicare claims for PET and CT scan of the chest, abdomen and pelvis performed after surgical resection were recorded and analyzed. In order to evaluate the use of definitive treatment (DT) for presumed recurrent disease, we identified patients who had undergone lung, liver, or colorectal resection during follow-up. Results: Of 22,544 patients who underwent primary resection of colon cancer, 13,753 (61.0%) underwent PET and/or CT after resection and 1,109 (4.52%) underwent at least 1 DT. The mean number of imaging performed for patients who did not undergo DT was 1.68. Within this group, the utilization of PET and/or CT varied by stage, with 14.1%, 22.4%, and 42.2% of patients with stage I, II, and III disease undergoing imaging, respectively. The most significant factor associated with PET and/or CT was tumor stage with an odds ratio of 1.86 (95% CI 1.70-2.02, P&lt;0.001) for stage II disease and 4.56 (95% CI 4.17-4.99, P&lt;0.001) for stage III disease compared to stage I disease. Additional factors significantly associated with PET and/or CT included year of diagnosis and SEER region. Among patients who did undergo DT, the mean number of PET and/or CT performed was 5.37 but the majority of imaging studies were performed 4 months prior to DT (mean 2.95). Conclusions: There is a high degree of variation in the use of surveillance imaging among elderly patients with localized colon cancer although some of this variation appears to be associated with stage at presentation. Our findings highlight discordance between clinical practice and surveillance guidelines. Further study is needed to understand the basis and appropriateness of clinical decision making that departs from guideline-based care.

Regional variation in surgical assessment of lymph nodes for staging among women with early-stage epithelial ovarian cancer

ObjectiveTo investigate geographical and socioeconomic variations in performance of lymph node dissection for the evaluation of patients with early-stage epithelial ovarian cancer. MethodsA population-based, retrospective cohort study was conducted using data from the National Cancer Institute's SEER Program for 15 geographic registries and county-level measures. Women with early-stage epithelial ovarian cancer registered between 2000 and 2008 with known lymph node assessment status were studied. A multiple logistic regression analysis was used to evaluate the differences in the likelihood of lymph node assessment according to geographic SEER region. ResultsAfter adjusting for tumor characteristics, demographics, and area-based socioeconomic measures, a significant relationship between SEER region and lymph node dissection remained. Compared to the region with the highest proportion of lymph node dissection, there is a significantly lower probability of surgical assessment of lymph nodes in 8 of the remaining 14 geographical regions. ConclusionsThe variation in ovarian cancer surgical care by region reported in this study has implications for access and outcomes for patients with early-stage disease. Study findings merit further investigation and should be characterized to permit targeted interventions aimed at reducing the observed disparities.

Evaluating comparative effectiveness with observational data

A previous observational study reported that endoscopic ultrasound (EUS) is associated with improved survival in older patients with pancreatic cancer. The objective of this study was to reevaluate this association using different statistical methods to control for confounding and selection bias. Surveillance, Epidemiology, and End Results (SEER)-Medicare linked data (1992-2007) was used to identify patients with locoregional pancreatic cancer. Two-year survival in patients who did and did not receive EUS was compared by using standard Cox proportional hazards models, propensity score methodology, and instrumental variable analysis. EUS was associated with improved survival in both unadjusted (hazard ratio [HR] = 0.67, 95% confidence interval [CI] = 0.63-0.72) and standard regression analyses (HR = 0.78, 95% CI = 0.73-0.84) which controlled for age, sex, race, marital status, tumor stage, SEER region, Charlson comorbidity, year of diagnosis, education, preoperative biliary stenting, chemotherapy, radiation, and pancreatic resection. Propensity score adjustment, matching, and stratification did not attenuate this survival benefit. In an instrumental variable analysis, the survival benefit was no longer observed (HR = 1.00, 95% CI = 0.73-1.36). These results demonstrate the need to exercise caution in using administrative data to infer causal mortality benefits with diagnostic and/or treatment interventions in cancer research.

Incremental cost (IC) analysis of skeletal related events (SREs) among elderly men with stage IV metastatic (M1) prostate cancer (PCa).

e16034 Background: Patients diagnosed with stage IV M1PCa are predisposed to SREs, such as pathologic fracture (PF), spinal cord compression (SCC) and bone surgery (BS). There is limited information in the literature regarding the ICs associated with SREs among stage IV M1 PCa patients. Methods: We analyzed patients aged 66+ yrs diagnosed with incident stage IV M1 PCa between 2000 and 2007 from the linked SEER-Medicare dataset. Five mutually exclusive SRE categories were created: PF only, PF+concurrent surgery (SRG), SCC only, SCC+SRG, and BS only. Patients with multiple SREs were excluded. A propensity score for the incidence of an SRE was estimated using a logistic regression model including baseline demographic and clinical variables, diagnosis year and SEER region. Patients with SREs (cases) were matched to those without SREs (controls), based on the propensity score. The date of SRE of a case was assigned to a matched control, and the 12 month pre- and post-SRE costs were calculated. A difference-in-difference method was used to estimate the ICs (post-pre) for cases vs. controls. The analysis was conducted from a US Medicare perspective. Results: Application of inclusion criteria resulted in 1,131 stage IV M1 PCa patients with SREs. The average age was 78 yrs and 12% were African American. Using the propensity score, 1,031 cases were matched with 1,031 controls and allocated to the following SRE groups: PF+SRG (n=134), PF only (n=143), SCC+SRG (n=40), SCC only (n=538) and BS only (n=176). The average IC per SRE was $30,548 (vs. controls). The most expensive SRE group was SCC+SRG with a total average IC of $62,412, followed by BS only ($37,554), PF+SRG ($35,520), SCC ($28,027), and PF only ($17,839). Inpatient costs were the major driver of ICs followed by physician/non-institutional provider and skilled nursing facility (SNF) costs. Conclusions: The ICs associated with an SRE are significant. The IC of an SRE varies by type of SRE and service category. [Table: see text]