Search

PurposeTo evaluate the association of subjective social status (SSS) with metabolic syndrome (MetS) severity and its potential contribution to racial health disparities in women with breast cancer. Methods Multi-center cross-sectional study (10 US hospitals) in women (n=1206) with primary diagnosis of invasive breast cancer received during Mar/2013- Feb/2020. Participants, self-identified as non-Hispanic white or black, underwent physical and laboratory examinations and survey questions assessing socioeconomic parameters, medical history and behavioral risks. SSS was measured with the 10-rung McArthur scale. MetS severity was measured with a validated Z-Score. Generalized linear mixed modeling was used to analyze the associations. Missing data were handled using multiple imputation. ResultsAverage age was 58 years. On average, the SSS of black women, given equivalent level of income and education, was lower than the SSS of white women: 6.6 (6.1 to 7.0) vs 7.7 (7.54 to 7.79) among college graduates and 6.8 (6.4 to 7.2) vs 7.6 (7.5 to 7.8) among women in the high-income category (> $75,000). In multivariable analysis, after controlling for age, income, education, diet and physical activity, increasing SSS was associated with a decrease in MetS-Z, - 0.10 (-0.16 to -0.04) per every 2 rung increase in the McArthur scale. Conclusion Black women with breast cancer rank their SSS lower than white women with breast cancer do at each level of income and education. As SSS is strongly associated with MetS severity these results identify potentially modifiable factors that contribute to racial disparities.

560 Background: Metabolic syndrome (MS) is associated with worse breast cancer prognosis. Black women have higher rates of advanced breast cancer, as well as MS, diabetes and obesity. As socioeconomic status is associated with MS, we asked whether the subjective perception of social status (SSS), might influence this association in black and white women with a new breast cancer (BC) diagnosis. Methods: We surveyed, obtained serum and conducted anthropometric measures of 1206 women with a new BC diagnosis. Triglycerides, systolic blood pressure (SBP), waist circumference (cm), HDL and glucose were used to calculate a severity index of MS (MS-Zscore: mean=0; sd=1). Women reported their SSS on the 10 rung McArthur US social status ladder, their household income, education attained, diet quality (5 point scale), and exercise measured with metabolic equivalents (METS). Data were analyzed with multivariable generalized linear models. Missing data were imputed with multiple imputation. Results: Average age was 58 yrs of 295 black and 911 white women. BC stage > II was in 11.6% of black and 2.4% of white women. On average black women had higher BMI (31.5 vs 26.6; pth %ile), and less likely to graduate college (30.3% vs 70.4%). MSZ-score was positively associated with age (.02 per year; p

e18142 Background: Much of cancer health services research relies on administrative data, yet, there have been calls for more granular racial & social disparities data. Subjective SES (sSES) is associated with health status & behaviors, independent of objective SES (oSES) measures like income, education and race. We assessed the relationship between breast cancer patients’ oSES and sSES with their cancer screening and health behaviors. Methods: Data are part of a large prospective study evaluating the role of insulin resistance in women with newly diagnosed breast cancer. Patients were asked to place where they stand in their communities on the MacArthur Scale of Subjective Social Status ladder (sSES). oSES measures include race, income & education. We assessed diet, activity, breast & pap screening. Group comparisons used chi-square and t-tests as appropriate. We ran logistic multivariate models with age, race, insurance, comorbidity & income. Results: Of 1035 breast cancer patients with an average age of 58±12 yrs, 81% were White and 19% Black, 63% graduated college, 47%W & 16%B women reported an income of > $100,000/yr. The median sSES = 8; 20% had high sSES ( > 8). There were no racial differences observed in sSES (high sSES: 22% B vs 20% W; p = 0.7). More patients with high sSES graduated college (76% vs 61%, p = 0.0002) & had income > $100K/yr (62% vs 36%, p < .0001) than patients with lower sSES. Of the oSES, women with higher income were more likely to undergo both cancer screenings as compared to patients with lower income. College education did not impact cancer screenings. Black women were less likely to get pap smears. Patients with high as compared to low sSES had higher cancer screening rates and healthy behaviors. Conclusions: Income and sSES are positively associated with cancer screening and health behaviors; education & race are associated with activity & diet. Race is associated with pap screening.[Table: see text]

e18157 Background: Black women are more likely to die of breast cancer and develop more aggressive subtypes than white women. Black women are also more likely to be obese and have insulin resistance than white women. Insulin resistance has been associated with faster tumor growth but has not been studied as a potential mediator of racial disparities in women with breast cancer. We hypothesized that black women would present with more aggressive breast cancer and this would be associated with obesity and insulin resistance. Methods: We recruited 1017 (80% white, 20% black) women with new primary breast cancer, measured fasting blood glucose and insulin, body mass index (BMI), triple negative breast cancer (TNBC) & Nottingham prognostic index (NPI). We classified aggressive breast cancer as NPI > 4.4. We calculated insulin resistance scores (HOMA) and classified insulin resistance as HOMA > 2.8. Patients self-identified race. Results: Of 1017 women, average age was 58 years (SD = 12.0). 373 (37%) were stage 2+ at time of diagnosis; 19% had an NPI > 4.4. Black women presented with higher stage of cancer than white women (stage 2+: 45% vs 35%; p = 0.01), more TNBC than white women (10% vs 5%, p = 0.01), were more insulin resistant (24% vs 11%, p < .0001), had higher BMI (31.4kg/m2 vs 26.6 kg/m2; p < .0001) and NPI > 4.4 (29% vs. 17%, p = 0.0002) than white women. HOMA score was positively but not significantly associated with NPI score (r = 0.05; p = 0.1). Multivariate mediation regression model suggested that HOMA_IR does not mediate the effect from black race to higher NPI score (β = 0.01; 95%CI: -0.017 to 0.039). Conclusions: In women with newly diagnosed breast cancer, black women are more likely to be obese, have higher HOMA & NPI scores than white women. While these data are consistent with the hypothesized relationship of hyperinsulinemia promoting more aggressive breast cancer, to date, insulin resistance does not appear to mediate the effect of race and poor prognostic breast cancer.

e18140 Background: Multiple Myeloma (MM) management has significantly improved disease-free and overall (OS) survival but disparities among racial groups still exist. After the Affordable Care Act, the extent to which induction, autologous stem cell transplant (ASCT), and maintenance therapies are used are uncertain. We sought to describe underuse of induction, ASCT and maintenance in a large referral center. Methods: Between 2010 and 2014, 3101 patients were diagnosed with MM via ICD-9 code from the Data Warehouse and certified hospital tumor registry. NCCN 2014 and CMS guidelines were used to define the categories of treatment underuse, and define transplant eligibility. Demographics including insurance, Charlson Comorbidity Index and treatments received were determined via chart abstraction. To date, 393 confirmed MM from 697 charts were abstracted. Comparison by groups used Chi-square for categorical variables, t-test and ANOVA for continuous variables. Multivariate logistic regression models were applied to predict underuse of induction, harvest, ASCT, and maintenance. Results: Patients were 62 ±11.3 years-old, with no racial differences in age and insurance coverage. More minorities had Medicaid (Black [B] 13%, White [W] 7%, Hispanic [H] 25%; p = 0.001). Almost all patients (97%) received induction (B 99%, W 96%, H 100%; p = 0.3), with no difference by insurance. Among transplant eligible patients, 93% underwent harvest, 87% underwent ASCT, with no racial differences. Patients with Medicare or self-pay were less likely to undergo harvest compared to patients with Medicaid or private insurance (p = 0.01). No difference in ASCT rates by insurance were noted. B patients were less likely to receive maintenance than non-B (73% vs 86%; p = 0.03), with no difference by insurance. OS was 73%, with no racial differences. In multivariate model, older age predicted induction underuse (aOR = 1.15, 95% CI: 1.06-1.25) (c = 0.9, p = 0.005), and B patients experienced more maintenance underuse (aOR = 2.22, 95% CI: 1.09-4.54) (c = 0.61, p = 0.1), controlling for age and comorbidity. Conclusions: While there were no racial or insurance differences in access to induction therapy, fewer Black patients received maintenance therapy. Interviews are underway to understand reasons for observed differences.

e18159 Background: Care coordination affects the quality of cancer treatment. Multiple Myeloma (MM) patients frequently are cared for by local and referral center physicians. We sought to determine whether getting treated at different sites might impact on the quality of MM care. Methods: We used NCCN 2014 and CMS guidelines to define categories of treatment quality and transplant eligibility. We defined high quality MM care as: Induction for all patients with ECOG < 4; harvest & transplant for patients completing induction without progression, with albumin > 2.8, bilirubin < 3 and age ≤77 years; maintenance for transplant ineligible patients completing induction with no progression & for transplant eligible patients completing transplant with no progression; and supportive care of receiving bone targeting agents, anticoagulation and infection prophylaxis for all patients receiving MM therapy. Patients who received care with both local and referral center physicians were considered to have received care at 2 sites. We abstracted 709 charts of patients with ICD-9 diagnosis of MM from 2010-2014 for demographics, site of care and treatments received. We compared groups with chi square and multivariate logistic regression models with age, insurance, race, comorbidity and sites of care to predict care quality. Results: Of the 709, 388 had active MM. On average, patients were 62 ±11 years. Overall, 70% received care at > 1 site. Referred patients had no differences by age, comorbidity index, insurance or race. Having > 1 care site was associated with higher rates of induction and harvest. Conclusions: Historically, hand-offs in care have been associated with poorer quality. However, in a referral center, hand-offs do not negatively affect induction and harvest care quality. Patients referred to a MM center are more likely to get high quality treatments for which they are referred, but referral does not ensure receiving the full gamut of needed MM treatments.[Table: see text]