Surveillance, Epidemiology, And End Results Sites Research Articles

Population-Based Brain Tumor Survival Analysis via Spatial- and Temporal-Smoothing.

In cancer research, population-based survival analysis has played an important role. In this article, we conduct survival analysis on patients with brain tumors using the SEER (Surveillance, Epidemiology, and End Results) database from the NCI (National Cancer Institute). It has been recognized that cancer survival models have spatial and temporal variations which are caused by multiple factors, but such variations are usually not “abrupt” (that is, they should be smooth). As such, spatially and temporally pooling all data and analyzing each spatial/temporal point separately are either inappropriate or ineffective. In this article, we develop and implement a spatial- and temporal-smoothing technique, which can effectively accommodate spatial/temporal variations and realize information borrowing across spatial/temporal points. Simulation demonstrates effectiveness of the proposed approach in improving estimation. Data on a total of 123,571 patients with brain tumors diagnosed between 1911 and 2010 from 16 SEER sites is analyzed. Findings different from separate estimation and simple pooling are made. Overall, this study may provide a practically useful way for modeling the survival of brain tumor (and other cancers) using population data.

Assessment of Voice Outcomes Following Surgery for Thyroid Cancer

An increasing number of surgeries are being performed for differentiated thyroid cancer (DTC). Long-term voice abnormalities are a known risk of thyroid surgery; however, few studies have used validated scales to quantify voice outcomes after surgery. To identify the prevalence, severity, and factors associated with poor voice outcomes following surgery for DTC. A cross-sectional, population-based survey was distributed via a modified Dillman method to 4185 eligible patients and linked to Surveillance, Epidemiology and End Results (SEER) data from SEER sites in Georgia and Los Angeles, California, from February 1, 2017, to October 31, 2018. Multivariable logistic regression and zero-inflated negative binomial analysis were performed to determine factors associated with abnormal voice. Participants included patients undergoing surgery for DTC between January 1, 2014, and December 31, 2015, excluding those with voice abnormalities before surgery. Abnormal Voice Handicap Index (VHI-10) score, defined as greater than 11. The VHI-10 is designed to quantify 10 psychosocial consequences of voice disorders on a Likert scale (0, never; to 4, always). A total of 2632 patients (63%) responded to the survey and 2325 met the inclusion criteria. With data reported as unweighted number and weighted percentage, 1792 were women (77.4%); weighted mean (SD) age was 49.4 (14.4) years. Of these, 599 patients (25.8%) reported voice changes lasting more than 3 months following surgery, 272 patients (12.7%) were identified as having an abnormal VHI-10 score, and 105 patients (4.7%) reported vocal fold motion impairment diagnosed by laryngoscopy. In multivariable analysis, factors associated with an abnormal VHI-10 score included age 45 to 54 years (reference, ≤44 years; odds ratio [OR], 1.49; 95% CI, 1.05-2.11), black race (OR, 1.73; 95% CI, 1.14-2.62), Asian race (OR, 1.66; 95% CI, 1.08-2.54), gastroesophageal reflux disease (OR, 1.67; 95% CI, 1.15-2.43), and lateral neck dissection (OR, 1.99; 95% CI, 1.11-3.56). A high prevalence of abnormal voice per validation with the VHI-10 emphasizes the need for heightened awareness of voice abnormalities following surgery and warrants consideration in the preoperative risk-benefit discussion, planned extent of surgery, and postoperative rehabilitation.

Robustness of RNA sequencing on older formalin-fixed paraffin-embedded tissue from high-grade ovarian serous adenocarcinomas.

Formalin-fixed paraffin-embedded (FFPE) tissues are among the most widely available clinical specimens. Their potential utility as a source of RNA for transcriptome studies would greatly enhance population-based cancer studies. Although preliminary studies suggest FFPE tissue may be used for RNA sequencing, the effect of storage time on these specimens needs to be determined. We conducted this study to determine whether RNA in archived FFPE high-grade ovarian serous adenocarcinomas from Surveillance, Epidemiology and End Results (SEER) registries was present in sufficient quantity and quality for RNA-Seq analysis. FFPE tissues, stored from 7 to 32 years, were obtained from three SEER sites. RNA was extracted, quantified, quality assessed, and subjected to RNA-Seq (a whole transcriptome sequencing technology). FFPE specimens stored for longer periods of time had poorer RNA sample quality as indicated by negative correlations between specimen storage time and fragment distribution values (DV). In addition, sample contamination was a common issue among the RNA, with 41 of 67 samples having 5% to 48% bacterial contamination. However, regardless of specimen storage time and bacterial contamination, 60% of the samples yielded data that enabled gene expression quantification, identifying more than 10,000 genes, with the correlations among most biological replicates above 0.7. This study demonstrates that FFPE high-grade ovarian serous adenocarcinomas specimens stored in repositories for up to 32 years and under varying storage conditions are a promising source of RNA for RNA-Seq. We also describe certain caveats to be considered when designing RNA-Seq studies using archived FFPE tissues.

Survival improvement by decade of patients aged 0–14 years with acute lymphoblastic leukemia: a SEER analysis

To evaluate treatment outcomes in children with acute lymphoblastic leukemia (ALL) over the past 3 decades, we assessed the survival of children with ALL in the Surveillance, Epidemiology, and End Results (SEER) database. Among 12,096 patients from 18 SEER sites diagnosed from 1981 to 2010, survival rates improved each decade from 74.8% to 84.5% to 88.6% at 5 years and from 69.3% to 80.9% to 85.5% at 10 years (P < 0.0001). For ages 10–14 years, 10-year survival increased by more than 20 percentage points to 75.3%, but for infants, it remained low at 54.7%. Improvements in survival rates were observed in both sexes, but survival rates were higher in girls than in boys. For ages 0–14 years during the 2001–2010 period, the 10-year relative survival rates were 87.8% in girls and 83.6% in boys (P < 0.01). Survival rates in child with ALL are expected to further improve with continuous advance in therapies such as targeted therapy and personalized therapy.

Anal Cancer Incidence and Survival: Comparing the Greater San-Francisco Bay Area to Other SEER Cancer Registries

The incidence of squamous cell carcinoma of the anus, anal canal, and anorectum (SCCA) has increased over time. However, there are still no national guidelines on screening for SCCA among high-risk populations. Providers at University of California, San Francisco have been at the forefront of providing anal dysplasia screening. To determine whether such a screening program allows for earlier detection of abnormalities and consequently, improves patient survival, we conducted an ecological study using data from the Surveillance, Epidemiology, and End Results (SEER) program to compare the San Francisco-Oakland catchment area (SF-O) to other SEER sites where routine screening has not been as accessible. Cox regression models were utilized to assess the impact of residing in the SF-O region, versus other SEER sites, on cause-specific mortality hazard. Logistic regression was used to determine if site was associated with the probability of having an in situ versus invasive tumor among SCCA cases. All analyses were stratified on calendar time (1985–1995 and 1996–2008) to compare differences pre- and post- highly active anti-retroviral therapy. Among SCCA cases, being reported by the SF-O registry was associated with a four fold higher probability of having an in situ tumor (rather than an invasive tumor) [95% CI: 3.48–4.61], compared to sites outside of California, between 1996 and 2008. Cases reported from the SF-O region between 1996 and 2008 had a 39% lower mortality risk than those reported from registries outside California (95% CI: 0.51–0.72). However, there was no decrease in the rate of invasive SCCA over this period. This is the first ecological study to evaluate whether access to anal cancer screening programs may help improve patient survival by allowing for earlier detection of lesions. Our results imply that routine screening programs may help detect SCCA at an earlier stage and thus, potentially impact patient survival.

Lung Adenocarcinoma Incidence Rates and Their Relation to Motor Vehicle Density

The temporal trend of adenocarcinoma incidence rates of the lung (ADL) has been reported to parallel the trend of nitrogen oxide (NOx) emissions in the United States. This study explores the geographic pattern of ADL incidence and its relation to motor vehicle density, the major indicator of NOx emissions before 1970. ADL incidence rates by counties were retrieved from the nine sites of Surveillance, Epidemiology, and End Results (SEER) Program for the period of 1973 to 1990. Motor vehicle densities by county in 1970 for these SEER sites were estimated according to the numbers of motor vehicle registration and the size of each county. Regression analysis was done with the data of motor vehicle density by counties. A dose-response pattern between motor vehicle density and ADL incidence was found. The risks are 136% and 68% higher for ADL and squamous cell carcinoma, respectively, for male residents living in areas with 937 motor vehicles per square mile, compared with those living in areas with about one motor vehicle per square mile. The R2 are as high as 0.805 and 0.504 in regressions for male incidence rates of ADL and squamous cell carcinoma, respectively, with the vehicle density values. As a comparison, the prostate cancer incidence rates did not show dose-response relationship with motor vehicle density. If the effect of NOx emissions on ADL incidence rates can be proved by further studies, the current standard of allowance of NOx emissions may need to be revised.

Examining the characteristics of unstaged colon and rectal cancer cases

Background: There are only three published studies that have examined unstaged cancers, including breast, prostate and colorectal cancer, but none have specifically examined which aspects of tumor staging were missing. The objective of the study was to identify which stage components were missing, to characterize and to identify predictors of unstaged cancer, and to evaluate prognosis of unstaged in colon and rectal cancer patients. Methods: Data on all colon and rectal cancer cases diagnosed during 1991 to 2002 were identified from the Surveillance, Epidemiology and End Results (SEER) database. Information included in the analysis encompassed socio-demographics; tumor size (T); number of lymph nodes (N); metastases (M); SEER summary stage; SEER sites; cancer treatment; month and year of diagnosis; and last date known alive. Results: The study included 128,418 colon and 44,616 rectal cancer patients. Overall, 5.1% of colon and 7.8% of rectal cancer patients were unstaged. Compared to staged cases, both unstaged colon and rectal cancer cases were more likely to be older, African American, female, diagnosed in the Northeast and South, and have unknown treatment (all p < 0.001). The stage component of M was found to be the factor missing most frequently which attributed most to being unstaged. Survival was significantly higher in unstaged and the overall cohort compared to distant staged colon and rectal cancer patients (both p < 0.001) except for patients ≥ 65 years old ( p > 0.99). Conclusions: Given that unstaged cases differ from staged, the findings have methodological implications in accounting for unstaged cases in epidemiological and health services research.

The use of tumor markers in testis cancer in the United States: A potential quality issue

Background Performance measurement is currently being implemented at various levels to address issues related to quality of care. Administrative data and cancer registries provide a measure of health care delivery (processes) at the population level. In testis cancer, the use of tumor markers is an important component of care and may represent a measurable quality indicator. We analyzed the use of testis cancer tumor markers across the United States. Methods Incident germ cell testis cancer cases between 1998 and 2002 were abstracted from the Surveillance, Epidemiology, and End Results (SEER) program. Tumor marker codes were categorized as performed and not performed. Bivariate and multivariable analyses were performed. Logistic regression was used to determine the independent association of demographic and clinical factors with the failure to measure or document tumor markers. Results In 4,742 testis cancer cases, performance of both alpha fetoprotein and human chorionic gonadotropin was recorded in 44.7% of cases, while performance of alpha fetoprotein, human chorionic gonadotropin, and lactate dehydrogenase together occurred in only 16%. Tumor marker use did not increase considerably between 1998 and 2002, and significant variation existed between SEER sites. In multivariable models, disease stage and the interaction between SEER site and year were significantly associated with tumor marker use. Conclusions We have identified low and variable rates of tumor marker use in testis cancer cases recorded in SEER. As pay-for-performance evolves, standardized metrics for recommended processes of care (e.g., measurement of tumor markers in patients with testis cancer) may be tied to reimbursement. In this context, failure to document the use of tumor markers will be interpreted as less than optimal care.

Does Ovarian Cancer Treatment and Survival Differ by the Specialty Providing Chemotherapy?

Chemotherapy for ovarian cancer is usually administered by medical oncologists (MOs) or gynecologic oncologists (GOs). GOs perform a broad spectrum of surgical and medical activities while managing a limited number of diseases; MOs specialize in the administration of chemotherapy but manage a broad array of diseases. We asked whether survival, treatment, and toxicity differed according to the type of specialist providing the chemotherapy after surgery. Using Surveillance, Epidemiology, and End Results (SEER)--Medicare data for patients 65 years old from 1991 through 2001 from eight SEER sites, we identified 344 patients with ovarian cancer who were treated with chemotherapy by a GO after surgery. Using optimal matching and propensity scores based on 36 characteristics, we matched these patients to 344 similar patients who were operated on and staged by the same type of surgeon but who received chemotherapy from an MO. MOs administered chemotherapy over more weeks than did the GOs (16.5 v 12.1 weeks, respectively; P < .0023), and MO patients had substantially more weeks that included chemotherapy-associated adverse events than GO patients (16.2 v 8.9 weeks, respectively; P < .0001). However, there was no difference in 5-year survival rate between the GO and MO groups (35% v 34%, respectively; P = .45). GO- and MO-treated patients who were closely matched on prognostic characteristics experienced very different rates of chemotherapy-associated adverse events and very different chemotherapy treatment styles by specialty type; however, their survival was virtually identical.

Epidemiology of primary hepatic malignancies in U.S. children

The epidemiology of primary hepatic malignancies in U.S. children is poorly characterized. We analyzed the incidence, mortality, and characteristics of primary hepatic malignancies in U.S. residents less than 20 years of age. Fatal primary hepatic malignancies in persons less than 20 years of age, between 1979 and 1996, were identified using the multiple-cause-of-death database (National Center for Health Statistics). Histologically confirmed primary hepatic malignancies occurring between 1973 and 1997 were identified using the Surveillance, Epidemiology, and End Results (SEER) database. Between 1979 and 1996, 918 primary hepatic malignancy deaths (average, 0.7/1,000,000/year) were reported nationally among persons less than 20 years of age; rates were higher among Asians and among foreign-born children. Between 1973 and 1997, 271 primary hepatic malignancy cases were reported to SEER among persons less than 20 years of age, of which 184 (67%) and 83 (31%) were hepatoblastoma and hepatocellular carcinoma, respectively. Among children less than 5 years of age, hepatoblastoma accounted for 91% of primary hepatic malignancy cases, whereas among those 15 to 19 years of age, hepatocellular carcinoma accounted for 87% of cases. Five-year survival for hepatoblastoma was 52%, compared with 18% for hepatocellular carcinoma. In the SEER sites, between 1973 and 1977 and 1993 and 1997, hepatoblastoma rates increased (0.6 to 1.2/1,000,000, respectively), while hepatocellular carcinoma rates decreased (0.45 to 0.29/1,000,000, respectively). In conclusion, histologically confirmed hepatocellular carcinoma was reported in children less than 5 years of age, also, where hepatoblastoma is the predominant primary hepatic malignancy. Hepatocellular carcinoma has worse survival rates than hepatoblastoma, and its incidence has not increased. Better maintenance of databases may provide information about associated factors behind this unexpected occurrence. (H epatology 2003;38:560-566.)

Is mastectomy overused? A call for an expanded research agenda.

Is mastectomy overused? A call for an expanded research agenda.

Annual cancer incidence rates for Hispanics in the United States: surveillance, epidemiology, and end results, 1992-1996.

The expansion of the Surveillance, Epidemiology, and End Results (SEER) program and the determination of annual population estimates by county level for different racial/ethnic groups since 1990 allow the calculation of annual cancer incidence rates for Hispanics. Incidence rates were calculated for 11 SEER areas representing 25% of the Hispanic population. Standard regression analyses of log-transformed rates were used to determine the trends of the rates. An important measure of the cancer burden among Hispanics is the rank order of their cancers. For Hispanic males, the five major cancers (in declining order) are prostate, lung and bronchus, colon/rectum, non-Hodgkin lymphoma, and stomach cancers. For Hispanic females, the top five cancers are breast, colon/rectum, lung and bronchus, cervix, and endometrial cancers. Another measure of cancer burden is their rates relative to white non-Hispanics. Hispanic males have rates greater than white non-Hispanic males for stomach (1.6 times greater) and liver and IBD cancers (2.2), whereas Hispanic females have greater rates for cervix (2.2 times greater), liver and IBD (2.0), stomach (2.1), and gallbladder cancers (3.3). Other measures of cancer burden include the trends in Hispanic rates. Hispanic males have significant declining trends for all sites, prostate cancer, and urinary bladder cancer, and an increasing trend for liver and IBD cancers. Hispanic females have significant declining trends for cervix and urinary bladder cancers. The SEER cancer incidence rates and trends provide a general overview of the cancer burden among Hispanics residing in the SEER sites. This type of information is critical for determining interventions to reduce the cancer burden among Hispanics in the United States.

Evaluation of Direct Medical Costs Related to Cancer

In recent years, market forces and political processes have generated growing interest in the economic costs of disease to the individual, institutions, and society. It has long been recognized that these costs include the direct cost of disease treatment and care, other costs that accrue to patients and their families as a result of disease, and economic losses to society as a whole (1). The publication of the article by Taplin et al. (2) in this issue of the Journal marks an important advance in empirical economic research on the first of these components, direct medical costs, related to cancer. Taplin et al. (2) are able to take advantage of a fortunate combination of existing data resources: the computerized internal cost-accounting system of Group Health Cooperative (GHC) of Puget Sound (Washington State) and the Cancer Surveillance System in the Seattle area, which participates in the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute (NCI). Using the linked GHC-SEER database, Taplin et al. (2) are able to present cost estimates related to cancer treatment at a level of detail and clinical relevance that could not be achieved in the past (3). These data support several types of analytical health economic studies that are increasingly important, including descriptive studies designed to understand aggregate annual levels of cost, average patient level lifetime costs (the flow of costs from diagnosis to death), and variations in these costs. While not reported in the current article, the GHC-SEER database also makes it possible to analyze which factors—treatment approach, service, and procedure utilization—drive costs. These data will also be useful in analytical studies designed to evaluate the cost-effectiveness of specific cancer prevention, screening, treatment, and care interventions. Data of this kind may also serve as a benchmark for studies designed to assess different organizational strategies for delivering cancer-related services. The richness of the GHCSEER database makes it possible to explore some aspects of cancer treatment in more detail than previously possible. For example, the variations in cost by age and comorbidity described in this article raise important questions regarding patterns of cancer care that deserve further investigation. In the past, when stage-specific costs were crucial to the analysis, as in the economic assessment of cancer-screening interventions or stage-specific treatment interventions, analysts often had to rely on anecdotal information or interpolations based on hypothetical clinical scenarios. Researchers have recognized the weakness of existing cost data and have stated the need for reliable stage-specific lifetime cost data (4). In response to this need, the NCI has sponsored a number of efforts. Two other NCI-sponsored projects also use linkage of financial and clinical data to SEER: a similar project involving Northern California-based Kaiser Permanente; and a link among all SEER sites and financial records of Medicare, maintained by the Health Care Financing Administration (5). Taplin et al. (2) provide estimates for both (not to be confused with lifetime) and net costs. What is the importance of this distinction? Total costs refer to all medical costs attributed to a patient diagnosed with cancer, regardless of whether a particular medical procedure or service was directly related to cancer treatment. Total cost estimates are much easier to construct but are not as useful as cost estimates for many analytical questions. Net costs, if estimated appropriately as they appear to be here, are a more accurate reflection of the costs avoided as a result of an intervention that prevents the occurrence of cancer, such as abatement of environmental or lifestyle risk factors. For the analysis of cancer screening, it may be appropriate to use either total or cost, depending on the perspective of the analysis and the analytical model that is being used.