Depression, chronic pain, and high-impact chronic pain among cancer survivors

12085

Background: The majority of the 17 million individuals living with a cancer diagnosis in the United States have experienced pain, either from the disease itself or from its treatment. Pain negatively impacts psychosocial quality of life and is associated with poorer overall outcome. However, the impact of pain on daily living differ among cancer survivors, and there is a paucity of research on chronic pain, especially high-impact chronic pain (HICP) in this growing population. We estimated the prevalence of chronic pain, and HICP among cancer survivors, and described the association between depression and these outcomes. Methods: This study used data from the 2015-2017 National Health Interview Survey. Outcomes of interest were chronic pain, defined as pain on most days or every day in the past six months, and HICP, defined as chronic pain that limited life or work activities on most days or every day during the past six months. Weighted, adjusted multivariable logistic regressions estimated association between depression and chronic pain and HICP among cancer survivors, while controlling for age, gender, marital status, education, employment, health insurance, smoking status, number of doctor’s visit, general health, and comorbidities. Results: Among 49,326 survey respondents, 11.7% (n = 5,335) had a cancer diagnosis. An estimated 43.6% of cancer survivors reported chronic pain; and 19.2% reported HICP. We found an association between depression and both chronic pain and HICP in unadjusted analyses. In the adjusted models, cancer survivors depressed within the last month had more than double the odds of reporting both chronic pain (aOR = 2.32; 95% CI 1.75, 3.07) and HICP (aOR = 2.12; 95% CI 1.50, 3.01). Other factors associated with both chronic pain and HICP among cancer survivors included being a current smoker (aORchronic pain = 1.63; 95% CI 1.14, 2.34; aORHICP = 1.83; 95% CI 1.18, 2.84) and being unemployed (aORchronic pain = 1.44; 95% CI 1.10,1.90; aORHICP = 3.10; 95% CI: 2.00−4.81). Cancer survivors with ≥2 comorbidities also had 55% increased odds of reporting chronic pain (aOR = 1.55; 95% CI 1.17,2.04) compared with those without comorbidities. Conclusions: Over 40% of cancer survivors may have a history of chronic pain, and survivors reporting being depressed are significantly more likely to report both chronic pain and HICP. The association between depression and pain in cancer survivors calls for personalized management of chronic pain, especially in cancer survivors with a history of depression.

Abstract A117: Thyroid cancer incidence trends among pediatrics, adolescents, and young adults in the United States 2001-2015

Background: The United States and other industrialized nations are experiencing what is known as the thyroid cancer epidemic. While there are studies of incidence trends among adult thyroid cancer patients in the United States, there is a paucity of data examining the relative contributions of different races/ethnicities and age groups, especially among the pediatric, adolescent, and young adult (AYA) population. This study aimed to identify which racial/ethnic groups may be driving the incidence trends of thyroid cancer among the pediatric and AYA population in the United States.

Methods: We used data from the joint National Program of Cancer Registries/Surveillance, Epidemiology, and End Results databases from 2001 - 2015 (n = 149,578), which covered the entire United States (50 states and the District of Columbia). We calculated age-adjusted incidence rates of thyroid cancer for patients from 0 - 39 years, and rates were stratified by age, race/ethnicity, and sex and presented per 100,000 person-years (PY). Overall rates for the entire study period, rate ratios (RR) with 95% confidence intervals (CI), and annual percent changes (APC) comparing the 2001 rate and 2015 rate were calculated for each group. Joinpoint regression estimated increases/decreases in age-adjusted incidence over time for each group through average annual percent changes (AAPC).

Results: The majority of thyroid cancer patients in this study were females (83%), white (60%), and between 15 - 39 years (98%). The overall incidence was 6.17 per 100,000 PY with an AAPC of 4.22 (p < 0.01). From 2001 to 2015, there was a 79% increase in age-adjusted incidence rate of thyroid cancer pediatric and AYAs in the United States (4.29 per 100,000 PY in 2001 vs. 7.69 per 100,000 PY in 2015). Both pediatric (0-14 years) and AYA (15 - 39 years) groups experienced significant increase in incidence of thyroid cancer in this time period (AAPC for 0 - 14 years = 4.38; AAPC for 15 - 39 years = 4.22; p < 0.01). The AYA population was 34 times as likely to develop thyroid cancer than the pediatric population (RR = 34.55, 95% CI 33.23, 35.94). Also, both males and females experienced significant rate increases from 2001-2015, but females were almost five times (RR = 4.81, 95% CI 4.74, 4.87) as likely to develop thyroid cancer than males (2.13 per 100,000 PY for males vs. 10.26 per 100,000 PY for females). Whites had the highest overall incidence rate (6.87 per 100,000 PY), while blacks had the lowest overall rate (2.89 per 100,000). Compared to whites, all other race/ethnicity groups were significantly less likely to develop thyroid cancer (RR range 0.42 - 0.90). All race/ethnicity groups had significant rate increases from 2001-2015 (AAPC range = 3.69 - 5.70).

Conclusion: Incidence of thyroid cancer increased in both the pediatric and AYA population over the last 15 years, and white females aged 15 - 39 years accounted for most of this increase. These findings can inform future directed screening and reveal unaddressed health disparities.

Citation Format: Katherine M. Polednik, Matthew C. Simpson, Aleksandr R. Bukatko, Matthew E. Gaubatz, Eric Adjei Boakye, Mark A. Vavaras, Nosayaba Osazuwa-Peters. Thyroid cancer incidence trends among pediatrics, adolescents, and young adults in the United States 2001-2015 [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr A117.

Abstract D128: Oropharyngeal cancer incidence-based mortality trends in the United States, 1985-2016

Objective: The last three decades in the United States have seen oropharyngeal cancer emerge as an important human papillomavirus (HPV)-associated cancer, with about three-quarters of cases thought to be positive for HPV. It has dramatically increased in incidence and recently surpassed cervical cancer as the leading HPV-associated cancer. While positive HPV tumor status generally portends better survival probability compared with non-HPV related head and neck cancer, there is a paucity of data describing mortality trends. This study aimed to describe trends in oropharyngeal cancer incidence-based mortality in the United States in the last three decades. Methods: We estimated age-adjusted incidence-based mortality rates (AAMR) from first primary oropharyngeal squamous cell carcinoma (OPSCC), using the Surveillance, Epidemiology, and End Results (SEER) 9 database from 1985-2016. To prevent later years from having a cumulatively larger set of patients diagnosed in the past, we only included OPSCC patients who died within 10 years of diagnosis. AAMRs were stratified by race, sex, and age at death and were presented per 100,000 person-years. Rate ratios (RRs) determined which groups had significantly different AAMRs, and Joinpoint regression calculated which groups had significant increases/decreases in annual AAMRs over time through annual percentage changes (APCs) and average APCs (AAPCs). We used 95% confidence intervals (CIs) to determine significant RRs, APCs, and AAPCs. Results: This study included 12,102 patients who died from first primary OPSCC from 1985-2016 with an AAMR of 1.16 per 100,000 person-years. AAMRs among males were 3.58 times higher than for females (RR = 3.58, 95% CI 3.43, 3.73). AAMRs among blacks were about 2 times higher than for whites (RR = 2.06, 95% CI 1.96, 2.16) but were about 60% lower for other race than whites (RR = 0.37, 95% CI 0.34, 0.42). From 1985-2009, AAMRs for first primary OPSCC decreased approximately 1.92% annually (APC = -1.92, 95% CI -2.27, -1.56) but remained stable from 2009-2016, which resulted in an average annual decrease of -1.31% from 1985-2016 (AAPC = -1.31, 95% CI -1.84, -0.78). When stratified by race or sex, all groups exhibited significant mortality rates decrease, however decrease was significantly greater among whites than blacks (white AAPC1985-2016 = -0.76; 95% CI -1.33, -0.17 vs black AAPC1985-2016 = -3.36; 95% CI -3.85, -2.87). AAMRs significantly decreased among 65+ year olds (AAPC = -0.88, 95% CI -1.63, -0.13), while AAMRs for 15-39 and 40-64-year olds exhibited non-significant decreases. Conclusions: While there has been significant decrease in oropharyngeal cancer mortality in the last three decades in the United States across age groups, races/ethnicity, and gender, there remained a significant mortality gap between blacks and whites, highlighting the persistent cancer-related disparity in the United States.

Citation Format: Nosayaba Osazuwa-Peters, Matthew C Simpson, Sean T Massa, Eric Adjei Boakye, Kara M Christopher, Sai D Challapalli, Katherine M Polednik, Haley N Bray, Greg M Ward, Mark A Varvares. Oropharyngeal cancer incidence-based mortality trends in the United States, 1985-2016 [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr D128.

Abstract A115: Socioeconomic disparities associated with 90-day mortality among patients with head and neck cancer in the United States

Background: There are previous studies on the impact of socioeconomic status on head and neck cancer outcomes, but it is not clear whether these factors are associated with short-term mortality as most studies on risk factors for mortality have focused on long-term mortality and clinical factors. This study aimed to quantify 90-day mortality rates and identify socioeconomic factors associated with 90-day mortality among patients with head and neck cancer.

Methods: This retrospective cohort study included 260,011 patients from the National Cancer Database (2004 to 2014) ≥ 18 years with a diagnosis of head and neck cancer and treated with curative intent with a combination of either surgery, radiation, and/or chemotherapy. Our outcome of interest was any-cause mortality within 90 days of first treatment. The effects of socioeconomic factors on 90-day mortality were estimated using the Cox proportional hazards model with the following adjustments: Heaviside function for time-varying effects and Šidák correction for familywise error (multiple comparisons). A multinomial cumulative logit model estimated the likelihood of higher comorbidity scores in variables of interest.

Results: There were 9,771 deaths within 90 days of treatment, yielding a 90-day mortality rate of 3.8%. Several socioeconomic factors were associated with 90-day mortality. Blacks (aHR = 1.10, 95% CI 1.00, 1.21) and males (aHR = 1.07; 95% CI 1.00, 1.15) were marginally more likely to die within 90 days of treatment. Hazard of 90-day mortality was significantly greater among patients who were uninsured (aHR = 1.71; 95% CI 1.48, 1.99) or insured by Medicaid (aHR = 1.72; 95% CI 1.53, 1.93) or Medicare (aHR = 1.40; 95% CI 1.27, 1.53), compared to those with private insurance. Residence in a zip-code with lower median income was associated with greater hazard of 90-day mortality [(aHR <$30,000 = 1.30; 95% CI 1.18, 1.44); (aHR $30,000 - $34,999 = 1.24; 95% CI 1.13, 1.36); (aHR $35,000 - $45,999 = 1.18; 95% CI 1.08, 1.27)], while farther travel distance for treatment was associated with decreased hazard of 90-day mortality [(aHR 50 - 249.9 miles = 0.86; 95% CI 0.77, 0.97); (aHR >250 miles = 0.70; 95% CI 50, 0.99)]. In addition, farther travel distance for treatment was associated with lower comorbidity scores [(aOR 50 - 249.9 miles = 0.91; 99% CI 0.86, 0.97); (aOR >250 miles = 0.78; 99% CI 0.67, 0.92)].

Conclusions: While the 90-day mortality rate was low among this national cohort of patients with head and neck cancer, there were significant sociodemographic disparities observed. Males, blacks, those uninsured, those with Medicaid or Medicare, and those living in poorer zip codes were more likely to die within 90 days of treatment, highlighting issues associated with access to care. To improve short-term head and neck cancer outcomes, these socioeconomic disparities associated with differing mortality rates among this cancer patient population need to be addressed.

Citation Format: Matthew E. Gaubatz, Aleksandr R. Bukatko, Matthew C. Simpson, Katherine M. Polednik, Eric A. Boakye, Mark A. Varvares, Nosayaba Osazuwa-Peters. Socioeconomic disparities associated with 90-day mortality among patients with head and neck cancer in the United States [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr A115.

Mental distress among adolescent and young adult (AYA) and adult cancer survivors

e19116

Background: Mental distress is associated with poor treatment adherence and adverse psychosocial outcomes, and cancer survivors, especially adolescent and young adults (AYA), may experience greater distress than older adults and the general population. We tested this hypothesis by examining the association between AYA vs. adult cancer survivors vs. the general population without a history of cancer, and mental distress. Methods: Using the 2014-2017 National Health Interview Surveys, 2,516 AYA cancer survivors (aged 15 – 39 years) were identified. We then used propensity score matching (matched on sociodemographics, comorbidities, smoking status and visit to mental health professional in past year) to create 2,516 older cancer survivors (aged ≥ 40 years); and 2,516 adults without cancer (general population) as the comparison groups. Mental distress (outcome of interest) was measured using the validated Kessler nonspecific mental/psychological distress (K6) scale. The 6-item K6 scale examines how frequently within the past 30 days an individual felt nervous, hopeless, restless or fidgety, worthless, sad, and that everything was an effort. Responses were summed to yield a score ranging between 0 and 24 and classified as none/low (0≤K6 < 5), moderate (5≤K6 < 13), and severe (K6≥13) mental distress. Two separate weighted multinomial logistic regression models estimated the odds of mental distress in study population (AYA vs. adult cancer survivors; and AYA vs. general population), adjusting for known covariates. Results: Mental distress was more prevalent among AYAs than adult cancer survivors (moderate: 24.0% vs 18.3%; and severe: 5.7% vs 4.2% [ P= .0002]); and the general population (moderate: 24.3% vs 16.7%; and severe: 6.1% vs 5.3% [ P< .0001]). Similarly, prevalence was higher among adult cancer survivors than the general population (moderate: 16.8% vs 13.6%; and severe: 3.2% vs 2.7% [ P= .0002]). In the multivariable multinomial analyses, AYAs had greater odds of mental distress (aORmoderate = 1.44; 95% CI 1.09, 1.89; and aORsevere = 1.77; 95% CI 1.21, 2.58) vs. adult cancer survivors. AYAs also had greater odds of mental distress vs. the general population (aORmoderate = 1.39; 95% CI 1.08, 1.79), but no significant difference in severe distress. Conclusions: About 1-in-4 AYA cancer survivors report some mental distress, and distress is more prevalent among this younger age group than older adults with cancer and the general population. Psychosocial care may be especially needed in this younger population to mitigate adverse psychosocial outcomes.

Cumulative odds of increased comorbid score in head and neck cancer

e17555

Background: Survival of head and neck cancer is impacted by known clinical factors, including anatomic subsite, stage of presentation, and treatment modality. An important clinical factor less explored is comorbidity burden. While it is known that a greater comorbidity burden is prognostic for poorer outcomes, it is unclear how the odds of presenting with increased comorbidity score is associated with head and neck cancer anatomic subsite. This study aimed at estimating the cumulative odds of increased comorbidity in head and neck cancer based on anatomic subsites. Methods: Data queried from National Cancer Database (2004-2015). Study sample (N = 328,504) consisted of Stage I-IV, Head and neck Squamous Cell Carcinoma (HNSCC) patients, with no missing demographic variables (age, sex, race, insurance status, local income, local population density). Multivariable cumulative logit model was used to estimate outcome of interest: odds of higher Charlson-Deyo comorbid condition score(CDCC) at HNSCC diagnosis. Results: Compared to patients diagnosed with oropharyngeal cancer (mostly HPV-related HNSCC), patients diagnosed with more tobacco-related HNSCC, such as laryngeal cancer (aOR: 1.69, 95% CI: 1.65-1.73), hypopharyngeal cancer (aOR: 1.33, 95% CI: 1.28-1.38), oral cavity (aOR: 1.26, 95% CI: 1.23-1.29), and sinonasal cancer (aOR: 1.12, 95% CI: 1.06-1.19) had greater odds of presenting with a higher CDCC. Patients with nasopharyngeal cancer did not statistically differ from oropharyngeal patients in odds of higher CDCC presentation. Conclusions: Patients diagnosed with cancers of larynx, hypopharynx and oral cavity (typically associated with tobacco and alcohol use) present with greater comorbid burden compared to patients diagnosed with oropharyngeal cancer (typically associated with HPV). It is important that the role of comorbidity burden be recognized in head and neck cancer prognostication.

Changes in the proportion of squamous cell carcinoma in head and neck cancer in the United States and Canada, 1995-2015

e17554

Background: There has been a shift in the epidemiologic landscape of head and neck cancer (HNC) with decreasing incidence of tobacco-related and increasing incidence of human papillomavirus (HPV)-related HNC. While it is often reported that ≥ 90% of HNC is considered squamous cell carcinoma (SCC), there is an apparent lack of recent population-based data to support this claim. This study aimed to estimate the current proportion and evaluate change in the proportion of SCC in HNC diagnoses in North America (United States and Canada) from 1995 to 2015. Methods: We queried the North American Association of Central Cancer Registries (NAACCR) database for HNC cases that were of either squamous (SQ) (ICD-0-3: 8050-8089) or squamous plus unspecified epithelial (SQE) (ICD-0-3:8010-8089) origin in the United States and Canada ( n = 1,054,409). All HNC included in the analysis were microscopically confirmed, malignant head and neck primary tumor sites of the oral cavity, nasopharynx, hypopharynx, oropharynx, nasal cavity, and larynx. Sub-analyses were conducted across more extensive cohort restriction combinations (country specific, registry specific, and primary sequence of cancer). Results: The overall proportion of SCC in HNC in North America from 1995-2015 was 81.7% (95% CI: 81.7 – 81.8) for SQ and 84.9% (95% CI: 84.8 – 85.0) for SQE. The proportion of SCC in HNC peaked in 2015 with 83.3% (95% CI: 83.0 – 83.6) for SQ and 85.9% (95% CI: 85.6 – 86.2) for SQE; and was lowest in 2005 with 80.7% (95% CI: 80.4 – 81.1) for SQ and 84.3% (95% CI: 83.9 – 84.6) for SQE. In the time period of this study (1995 – 2015), there were no years for which SQ or SQE made up 90% or more of HNC for any of the HNC cohorts. Conclusions: The changing landscape of HNC risk factors in the United States and Canada warrants re-evaluation and update of HNC epidemiological literature with regards to the proportion of SCC in HNC.

Factors associated with receipt of palliative-only care among survivors with head and neck cancer

73

Background: Head and neck cancer (HNC) survivors with end-stage disease may receive multimodal treatment with non-curative intent. However, besides clinical indications for palliative care, there is a paucity of data describing nonclinical factors associated with receipt of palliative care. This study aimed at characterizing sociodemographic factors associated with HNC survivors receiving palliative-only care. Methods: We used data from the National Cancer Database from 2004-2014, restricting our cohort to adult HNC survivors receiving palliative-only care. Multivariate logistic regression estimated association between sociodemographic factors and receiving palliative-only care. Survival of palliative-only patients was assessed via Cox proportional hazards model. Final regression models were adjusted for clinical/nonclinical covariates, including: age, comorbidity score, tumor site, cancer stage, histology, HPV status, and population density/urbanization. Results: Out of 325,489 HNC survivors, 2,404 received palliative-only treatment. Mean age was 69.1 years, and median survival was 4.9 months. Sociodemographic factors associated with receiving palliative-only care were gender: (female aOR=1.11; 95% CI 1.02, 1.22), race/ethnicity: [(Hispanic aOR=0.77; 95% CI 0.62, 0.97), (Non-Hispanic Black aOR=1.35; 95% CI 1.20, 1.52)], insurance status: [(uninsured aOR=2.59; 95% CI 2.13, 3.14), (Medicaid aOR=3.11; 95% CI 2.67, 3.61), (Medicare aOR=1.71; 95% CI 1.50, 1.95)], travel distance for care: (≥50 miles aOR=0.64; 95% CI 0.55, 0.74). Additionally, survivors receiving palliative-only care who were uninsured (aHR=1.34; 95% CI 1.08, 1.65), under Medicaid (aHR=1.29; 95%CI 1.10, 1.52), or Medicare (aHR=1.16; 95%CI 1.01, 1.33) were significantly more likely to die compared with those privately insured. Conclusions: Females, blacks, and those uninsured, under Medicaid/Medicare disproportionately contribute to the palliative-only population among HNC survivors. Disparities persist even at end-stage disease among HNC survivors and impact disease outcome.

Radiation and Second Primary Thyroid Cancer Following Index Head and Neck Cancer

OBJECTIVES/HYPOTHESIS

Radiation is thought to increase risk of developing second primary thyroid cancer (SPTC). This study estimated the rate of SPTC following index head and neck cancer (HNC) and determined whether radiation treatment among HNC survivors increased SPTC risk.

STUDY DESIGN

Retrospective data analysis.

METHOD

The Surveillance, Epidemiology, and End Results database (1975-2014) was queried for cases of index HNC (N = 127,563) that developed SPTC. Adjusted multivariable competing risk proportional hazards model tested risk of developing a SPTC following index HNC. Sensitivity analyses using proportional hazards models were also performed restricting data to patients who 1) received both radiation and chemotherapy and 2) radiation alone.

RESULTS

Only 0.2% of index HNC survivors (n = 229) developed SPTC, yielding a rate of 26.1 per 100,000 person-years. For every increasing year of age at diagnosis, patients were 3% less likely to develop an SPTC (adjusted hazard ratio [aHR] = 0.97, 95% CI: 0.96-0.98). Males were also less likely to develop an SPTC (aHR = 0.73, 95% CI: 0.55-0.96). Radiation (aHR = 0.92, 95% CI: 0.68-1.25), surgery (aHR = 0.79, 95% CI: 0.56-1.11), and chemotherapy (aHR = 1.13, 95% CI: 0.76-1.69) were not significantly associated with developing SPTC. The sensitivity models also did not find an association between treatment and risk of SPTC.

CONCLUSIONS

Rate of developing SPTC following index HNC was very low, and previous exposure to radiation did not significantly increase risk in our study population. More studies are needed to understand the increasing incidence of thyroid cancer across the United States.

LEVEL OF EVIDENCE

NA Laryngoscope, 129:1014-1020, 2019.

Effects of Essential Oil from Thymus vulgaris on Viability and Inflammation in Zebrafish Embryos

Innate immunity provides the initial response against pathogens and includes the inflammatory response. Regulation of the initiation and duration of neutrophil and mononuclear cell influx during inflammation determines both the successfulness of pathogen elimination and the level of resulting tissue damage. Zebrafish embryos provide excellent opportunities to visualize the inflammatory response. Neutrophil granules may be stained with Sudan black, and variation in neutrophil counts may be used to monitor the level of the response. Inflammation may be triggered by injuring the caudal fin, providing an opportunity for testing possible anti-inflammatory compounds in a whole-animal system. The use of homeopathic compounds as anti-inflammatory treatments is common in alternative medicine. Effects of unfractionated essential oil from Thymus vulgaris and its specific component, carvacrol, have been examined in cells in culture and in rodents. Our work extends this research to zebrafish, and includes toxicity and morphological studies as well as examination of anti-inflammatory effects following tail fin injury. Our results show that zebrafish are more sensitive to thyme oil compared to cells in culture, that cardiac defects arise due to thyme oil treatment, and that thyme oil reduces neutrophil infiltration during an inflammatory response.