Session

Recent Findings in Cancer Epidemiology - Poster Session

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Abstract

Trends in Endometrial Cancer Incidence among Adolescents and Young Women

Ifeoma Ozodiegwu, MPH1, Michele Cote, Ph.D2, Julie Ruterbusch, MPH2, Daniel Owusu, DrPH1, Megan Ashley Quinn, DrPH1 and Muyiwa Ategbole, Masters in Public Health1
(1)East Tennessee State University, Johnson City, TN, (2)Wayne State University, Detroit, MI

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Background:Endometrial cancer is the leading gynecologic cancer in the United States. Though it is typically a disease of older women, 2 – 14% of endometrial cancer cases occur in young women below 50 years. There is evidence of a rising trend in endometrial cancers for all women but variations between age groups have been poorly explored. This study describes trends in endometrial cancer incidence between younger and older women. Methods: Cancer cases with primary site listed as “Corpus and Uterus NOS” and diagnosed from 2000 to 2012 were selected from the National Cancer Institute Surveillance, Epidemiology and End Results (SEER) database which holds data for 18 population based registries in 14 states and covers 28% of the U.S. population. We split the 133,929 patients into a younger cohort aged 15 to 39 years (4,618) and an older cohort of women 40 years and older (129,311). Annual percentage changes (APC) and 95% confidence intervals were calculated by age group, race and cancer histology from age-adjusted yearly trends in incidence rates using SEER Stat. The APC was computed at a significance level of 0.05 using the weighted least squares regression method. Confidence intervals were derived with the Tiwari technique. Incidence rates were age-standardized to the 2000 US Standard Population. Results: The annual percentage change in incidence for younger women (APC=4.0%; CI, 3.0% - 4.9%) for the specified time period was approximately 7 times that of older women (APC =0.6; CI, 02% - 1.1%). Among racial groups, annual percentage change in incidence was significant in young white women (APC = 4.3%; CI: 3.3% - 5.4%), older black women (APC: 2.2%; CI: 1.5% - 2.9%), and older Asian or Pacific Islander women (APC: 2.4%; CI: 1.8% - 3.0%). Endometrioid histology subtypes were increasing in younger women (APC: 3.8%; CI: 2.7% - 5.0%). while poorer prognostic type II subtypes were increasing in older women (serous carcinomas - APC: 3.7%; CI: 2.8% - 4.7%; malignant mixed Mullerian tumors – APC: 2.1%; CI: 1.5% - 2.7%). Conclusion: The rate of new endometrial cancer cases appears to be growing faster in adolescents and young women than older women. Additional research into the complex interactions between race, age and risk factors such as obesity is warranted in order to mitigate the increasing rates of endometrial cancer in the US.

Epidemiology

Abstract

Relationship between social vulnerability, mammography, and breast cancer incidence and mortality: An epidemiological analysis of counties in the United States

Cataria Davis, MS, MPH, CHES1 and Justin McDaniel, MBA, PhD Candidate2
(1)SOUTHERN ILLINOIS UNIVERSITY, Carbondale, IL, (2)Southern Illinois University Carbondale, Carbondale, IL

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Introduction/Objectives: Socioeconomic status (SES) has been shown to predict breast cancer (BC) incidence and mortality. The Social Vulnerability Index (SVI), recently developed by the Hazards and Vulnerability Research Institute (HVRI) at the University of South Carolina, has the potential to provide a comprehensive measure of SES at the county level. We explored the relationship between the SVI (a potential diagnostic tool) and (a) breast cancer incidence/mortality and (b) mammography. Methods: County level data (N = 209) were downloaded for (a) the SVI from the HVRI, (b) BC incidence and in situ from the National Cancer Institute, (c) BC mortality from the Surveillance, Epidemiology, and End Results program, and (d) mammography screening rates from the Behavioral Risk Factor Surveillance System. Bootstrapped OLS regression models were calculated using the SVI as a predictor. Results: SVI significantly predicted mammography, (F = 4.82, p = 0.029, b = -0.38), BC incidence with a second order polynomial, (F = 25.18, p < 0.001, b = -2.94, b2 = -0.20), BC in situ, (F = 51.571, p < 0.001, b = -1.77), and BC mortality, (F = 4.38, p = 0.043, b = 0.47). Conclusion: The least vulnerable counties had the highest mammography, BC incidence, and BC in situ rates. These counties may have greater access to healthcare and, thus, physicians available to detect BC with early mammography. However, BC mortality was greatest among the most vulnerable counties; perhaps these counties have inadequate access to healthcare services and are unable to detect BC with early mammography.

Epidemiology Other professions or practice related to public health Public health or related research Social and behavioral sciences

Abstract

A Descriptive Study of a Cancer Cohort from a Regional Cancer Center in Rural Kentucky

Ashley Golden, PhD1, Martin D. Barrie, PhD, JD2 and Sadie Hutson, Ph.D., RN, WHNP3
(1)ORAU, Oak Ridge, TN, (2)Oak Ridge Associated Universities, Oak Ridge, TN, (3)The University of Tennessee, Knoxville, TN

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

The purpose of this study was to explore the most common cancers observed at a regional cancer center in rural Kentucky, with specific hypotheses: (1) earlier than expected age at diagnosis; (2) longer than expected survival for late stage malignancies; (3) higher than expected incidence of rare conditions; and (4) high prevalence of multiple malignances. Data were obtained from the Kentucky Cancer Registry (KCR) for the period 2003-2014. Cancer diagnoses were identified based on the Surveillance, Epidemiology, and End Results Program (SEER) classification. KCR did not begin reporting for triple negative breast cancer until 2010. All descriptive analyses and statistical comparisons were computed using SAS 9.3. There were 4,247 cancer cases representing 20 different malignancies and 4,048 patients. The top malignancies included, in order: lung and bronchus; breast; colorectal; prostate; melanoma of the skin; female genital organs; and urinary bladder. The youngest median age at diagnosis (59 years) was observed for breast cancer and female genital organ cancers. There were no apparent longer than expected survival for late stage malignancies. Over half of all Stage IV malignancies were lung and bronchus. There were 26 triple negative breast cancer cases, with 60 % (n=15) diagnosed before age 55. Other rare malignancies were identified: mesothelioma (3), Karposi Sarcoma (1), and sarcoma (5). Further investigation is warranted to explore and evaluate occurrences of rare malignancies and to characterize associations with family history, occupation, genetics, and survival. This information may aid in the development and expansion of prevention and early diagnosis interventions.

Clinical medicine applied in public health Epidemiology Planning of health education strategies, interventions, and programs Provision of health care to the public Public health or related nursing Public health or related research

Abstract

Cancer in First and second or subsequent generations Middle Easterners compared to Non-Hispanic Whites in California, 1988-2012

Clara Ziadeh and Hoda Anton-Culver, PhD
University of California, Irvine, Irvine, CA

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Background: Although Middle Easterners (ME) constitute one of the growing migrant populations in the United States, they are often mixed with the white ethnic group and therefore not well studied. Statistics showed that cancer incidence is lower in the Middle East compared to the United States. Unfortunately little is known of the difference in cancer incidence between ME living in the US and Non ME Non-Hispanic Whites (NHW) born in the US. The aim of this study is to compare the Proportional Incidence Ratio (PIR) between different generations of ME immigrants and NHW, stratified by gender. Methods: We used California Cancer Registry from 1988 to 2012 to identify all invasive cancer cases. Three different sample groups were created: First generation ME immigrants born in one of 22 ME countries and have a middle eastern last name, second or subsequent generations ME with middle eastern last name but born in the US and NHW who don't have a middle eastern last name, born in the US and from the white ethnic group. The cancers with the highest incidence in both females and males were selected and only the first cancers occurred were studied. PIR was calculated for the two ME groups with using NHW as the reference. Results: A total of 394,324 women and 417,657 men with invasive first cancers, were identified. ME first generation men had greater proportions of Prostate (6%), Colorectal (19%), Bladder (54%), Non-Hodgkin Lymphoma (17%), Leukemia (37%) and Stomach (123%) cancers. ME first generation women had greater proportions of Breast (16%), Colorectal (22%), Thyroid (92%), Leukemia (38%) and Stomach (220%) cancers. However, ME second or subsequent generations men had higher proportion of only Prostate (12%) cancer and women exclusively had a higher proportion of Thyroid (40%) cancer. Conclusion: Interestingly, the cancer proportion differences that we observed in the first generation of immigrants, almost disappeared in the following generations suggesting the importance of the acculturation process which can help understanding the cancer risk in different generations of immigrants.

Epidemiology

Abstract

Difference in MUC 4 Protein Expression and Phenotypic Characteristics of Pancreatic Cyst in Comparison to Pancreatic Cancer in Egypt

Asserewou Etekpo, Master of Public Health/Epidemiology1, Ghada Soliman, MD, PhD, RD, LMNT1, Audery J. Lazenby, Professor2, Ahmed Hablas, MD Surgical Oncology3, Baojiang Chen1, Surinder Batra, Professor2 and Amr Soliman, MD, PhD4
(1)University of Nebraska Medical Center, Omaha, NE, (2)University of Nebraska Medical Center College of Medicine, Omaha, NE, (3)Gharbia Cancer Society, Mansoura, Egypt, (4)University of Nebraska Medical Center College of Public Health, Omaha, NE

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Background: Pancreatic cancer is the fourth leading cause of cancer-related deaths in the United States with one year relative survival rate of 20% and five year rate of 6%. It is diagnosed at advanced disease stages where treatment is not effective. There are no biomarkers that have been shown to be sufficiently sensitive and specific to be applied to the general population. Thus searching for biomarkers for early detection is urgently needed. Methods: Immunohistochemistry assay was performed on paraffin embedded tissues of Pancreatic Cancer and Pancreatic Cyst from Egyptian patients to determine the expression of MUC4 protein. Logistic regression models were performed to predict the expression of MUC4 protein based on the pancreatic cancer, pancreatic cyst and type of cyst. Results: The analysis shows that patients with pancreatic cysts were younger compared to pancreatic cancer patients (Mean age of 28.7± 5.25 years vs 54.84±10.60) p=0.0001. The expression of MUC4 was not different between pancreatic cancer group and pancreatic cyst group (p=0.16). However; type of pancreatic cyst is a predictor of MUC4 expression. Mucinous and Serous cystadenoma cysts (80% vs 75%) express MUC4 protein 18 times more than the Pseudo cyst (0%) and Not Otherwise Specified cysts (25%) p= 0.022. Conclusions: This study showed an association between type of pancreatic cysts and MUC4 expression. It suggested that some pancreatic cyst patients are at risk of developing malignancy and need to be closely monitored. Future research will investigate these cyst patients to evaluate possible development of pancreatic cancer.

Chronic disease management and prevention Epidemiology

Abstract

Physical Activity and Breast Cancer in South Carolina: A Case-Control Analysis

Chisom Onyeuku, MSPH1, Swann Adams, PhD, James R. Hebert, MSPH, ScD3, Jan Eberth, PhD, Jiajia Zhang, PhD3 and Daheia J. Barr-Anderson, PhD, MSPH4
(1)Georgia Department of Public Health, Atlanta, GA, (2)University of South Carolina, Columbia, SC, (3)University of Minnesota, Minneapolis, MN

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

introduction: This study examined the relationship between breast cancer (BrCa) incidence and a priori recreational physical activity (RPA). This investigation was a secondary analysis of the Palmetto Women's Health Study, a cohort study conducted from 2000-2006 that examined how lifestyle and diet factors contributed to BrCa. methods: METs for lifetime and recent RPA were ascertained and reported in quartiles. Cases were confirmed by pathological comparison to the local tumor registry. Controls were matched by location and time. Analyses included multivariate logistic regression, chi-square and t-tests. Race, BMI and menopausal status were checked for effect modification. results: Cases tended to be older, reported more RPA and have higher BMI than controls (p<0.05). There was no significant difference in the odds of BrCa for any physical activity quartile in the analysis of recent RPA. There was a significant interaction between lifetime RPA and race (p=0.04). Non-Hispanic Black women who were at least 20 years old had a higher risk of BrCa when the 4th quartile was compared to the 1st: OR=37.36 (95% CI 3.71-372.93). In women age 65 and older, RPA had a protective effect on BrCa odds when comparing the 2nd and 4th quartiles to the 1st: OR=0.31 (95% CI 0.11-0.89) and OR=0.25 (95% CI 0.11-0.89), respectively. conclusions: These findings suggest that RPA may modify the odds of BrCa for younger, non-Hispanic Black women and all women who are above 65. This study has potential to aid health disparities research in the areas of both breast cancer and physical activity.

Chronic disease management and prevention Epidemiology

Abstract

Incidence of childhood cancer by type between 1999 and 2011 in Michigan

Jeremy Blankenship, MPA1, Gergana Kodjebacheva, PhD1, Lenwood Hayman Jr., Ph.D.1 and Shandowyn L. Parker, PhD, MPH2
(1)University of Michigan - Flint, Flint, MI, (2)University of Michigan-Flint, Flint, MI

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Background. Limited research investigates the incidence of leading childhood cancer types in Michigan. Procedure. The Center for Disease Control and Prevention WONDER database contains cancer incidence data. Using WONDER, we compared the age-adjusted incidence rates of cancer types by gender, race/ethnicity, and year among children and adolescents aged 0 to 19 years in Michigan and the United States. Results. Michigan had higher rates than the United States for 8 out of top 10 cancer types. Whites had higher incidence rates than Blacks for 9 of the 10 leading childhood cancers in both Michigan and the United States. Both Michigan Whites (189.6 per 100,000) and Michigan Blacks (132.3 per 100,000) had higher incidence rates than their United States counterparts (180.9 per 100,000 and 127.1 per 100,000 respectively). Michigan Blacks had higher incidence rates than their United States counterparts for six cancer types (Hodgkin's lymphoma, central nervous system and miscellaneous intra-cranial and intra-spinal neoplasms, astrocytoma, systematic nervous system tumors, soft tissue sarcomas, and germ cell, trophoblastic and gonadal neoplasms). Michigan Whites had higher incidence rates for four cancer types (I leukemia, II (a) lymphoma, III (b) astrocytoma, and X germ cell, trophoblastic and gonadal neoplasms) than United States Whites. Conclusions. Trends in childhood cancer incidence statistics tended to be worse in Michigan than in the United States. More awareness on the geographic differences in childhood cancer is needed. We propose research on the reasons for the cancer incidence rates in Michigan.

Epidemiology Public health or related research

Abstract

Factors associated with surgical acceptance of early stage non-small cell lung cancer

Poppy Deviany, MPH, Apar Ganti, MD, MS, FACP and Monirul Islam, MD, PhD
University of Nebraska Medical Center, Omaha, NE

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Background: Lung cancer is the leading cause of cancer death in the US. Improvement in treatment acceptance is expected to prevent more deaths of lung cancer. The present study examined factors associated with early stage non-small cell lung cancer (NSCLC) patients' acceptance of recommended treatment. Method: This study examined lung cancer patients recorded in the National Cancer Data Base diagnosed between 2003 and 2012. Analysis of treatment acceptance was carried out on surgical therapy for stage I and II NSCLC cases. We used logistic regression to examine factors associated with treatment acceptance. Results: The total cohort was 257,811 patients. Surgery was recommended to 187,674 (73%) patients, and 97% of them underwent surgery. Our multivariable analysis of these 187,674 patients showed that patients diagnosed at <50 years of age were more likely to accept surgery [AOR=2.72, 95% CI 2.06-3.59] compared to older patients. Other factors that were associated with higher acceptance of surgery included smaller tumor size (<3 cm) [AOR= 1.19, 95% CI 1.02-1.40], private insurance [AOR=2.47, 95% CI 1.89-3.23], and treatment at academic facilities [AOR=2.24, 95% CI 2.01-2.49]. Interestingly, we found that patients with comorbid conditions were more likely to undergo surgery [AOR=1.57, 95% CI 1.45-1.69] compared to those without comorbidity. Conclusions: Younger age, smaller tumors, private insurance status and treatment at academic facilities were associated with acceptance of surgery as treatment for early stage NSCLC. Further research into these is needed in order to design effective patient-centered approaches to increase patients' acceptance of recommended treatment and ultimately improve outcomes.

Epidemiology Public health or related research

Abstract

Oral cancer trends of age-adjusted incidence and mortality rates among African-American Adults in New Jersey between 1996-2012

Rufus Caine Jr., DDS MPH1, Khyati Mehta, BDS MPH2, Niyati Patel, Doctoral candidate3 and Shrey Shah, BDS3
(1)Rutgers University, School of Dental Medicine, 110 Bergen Street, Newark, New Jersey 07103-2425, Newark, NJ, (2)Rutgers University Medical School, Newark, NJ, (3)Rutgers School of Public Health, Newark, NJ

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

METHODOLOGY: Population-based data utilized a cross-sectional study design for 2,888 subjects obtained from NJSCR for 1996-2012. Sampling design was a non-probability based convenience approach with study groups including black males (BMs) and black females (BFs) 18 years and older with OCP. Bivariate logistical regression analyses were performed for APC with IBM SPSS Statistics Version 20 and Microsoft Excel 2010. RESULTS: There was a gradual decline in APC noted for all groups for AAIRs and AAMRs. The largest declines in APC for AAIR for BMs were salivary gland (-100.00) and floor of the mouth (-100.00); while largest increase was in nasopharynx (1.891). The largest declines in APC for AAIR for BFs were lip, floor of the mouth, nasopharynx, hypopharynx and tonsil (-100.00 for each). The largest increases in the AAIRs for BFs were tongue (1.251) and gum and other mouth (0.472). There were no increases in APC for AAMRs for all groups. The largest declines in APC for AAMR for BFs were tongue, floor of the mouth; and gum and other mouth (-100.00 each). CONCLUSION: The head and neck cancer incidence and mortality rates vary by gender and anatomical sites for African American adults disproportionately. Over 80% of OCP cases are directly attributable to alcohol and tobacco exposure. About 15-20% to non-smokers and non-drinkers; and 25% are linked to Human Papilloma Virus Type 16, and Human Immunodeficiency Virus. The increased susceptibility for the specific anatomical sites in these groups will direct us towards applying primary and secondary public health intervention strategies.

Epidemiology Program planning Public health or related research

Abstract

Trends in late-stage colorectal cancer incidence rates in Florida, 2000-2012

Korede Adegoke, MBBS, MPH, PhD1, Claudia Aguado Loi, PhD, MPH, CHES2, William M. Sappenfield, MD, MPH3, Clement K. Gwede, PhD, MPH, RN4 and Carol A. Bryant, PhD2
(1)College of Public Health , University of South Florida, Tampa, Tampa, FL, (2)University of South Florida, Tampa, FL, (3)University of South Florida, College of Public Health, Tampa, FL, (4)Moffitt Cancer Canter and Research Institute, Tampa, FL

APHA 2016 Annual Meeting & Expo (Oct. 29 - Nov. 2, 2016)

Objectives: Colorectal cancer (CRC) remains the 2nd leading cause of cancer-related deaths in the United States. With the documented increase in CRC screening, we hypothesize a decline in the rates of incident late-stage CRC. In this study, we examined the temporal trends in age-adjusted incidence rates of late-stage colorectal cancer among adults age 40-79 years in Florida from 2000-2012. Methods: Data from the population-based and statewide cancer registry, Florida Cancer Data System (FCDS) from 2000-2012 were analyzed (n=92,904). The outcome, late-stage of CRC was defined as stage III or IV according to the TNM staging system of the American Joint Committee on Cancer, 7th Edition. Age-adjusted rates of late stage CRC by sex, race and ethnicity were calculated for each year and adjusted to the 2000 US Census population. Trends in CRC rates were plotted and Mantel-Haenzsel tests for linear trends were conducted. Results: In this study, 43.8% (n= 40,733) of individuals with CRC age 40-79 years were diagnosed with late-stage CRC. Age-adjusted incidence rates of late-stage CRC decreased significantly from 40.7 to 30.4 per 100,000 persons from 2000- 2012 (p=0.001). Adjusted rates were consistently highest among males and blacks, and lowest among ‘other race' group. Irrespective of gender, a roughly equal decline (25%) in CRC incidence rate was observed. ‘Other race' experienced a 2.6% increase in late-stage CRC from 2000-2012. Blacks experienced minimal reduction in late-stage CRC rates (10.1%). Among those 40-49 years, a 16.6% significant increase in CRC rate was observed (p=0.025). A disproportionate reduction of late-stage CRC rates was observed among individuals 50-59 years of age as compared to those 60-69 and 70-79 years (15.7%, 32.6% and 37.7%, respectively). Conclusions: Findings highlight that certain subgroups have not experienced a decline nor equal reductions in late-stage CRC rates. Study implications demonstrates that these subgroups are not equally benefiting from current early detection efforts. Although cancer diagnosis at a younger age has been associated with more aggressive forms of cancer, special attention to these subgroups is warranted as documented CRC screening disparities have been seen among younger individuals within the recommended CRC screening age and non-whites. Understanding the unique factors impeding utilization and uptake of screening (including diagnostic colonoscopy) among these subgroups, as well as increasing access to screening, will be necessary to reduce late-stage CRC rates disparities.

Chronic disease management and prevention Epidemiology Public health or related research