CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

year-old age-group as prior validation studies indicated that ~94% of KS cases in this age-range are HIV-related. We calculated adjusted incidence rates and assessed annual trends among sociodemographic and geographic subgroups using joinpoint regression analysis. Results: During the study period, 11,712 men were diagnosed with KS. The overall incidence of KS among men decreased from 1.42/100,000 in 2000 to 0.92/100,000 in 2013, decreasing by 3.60% (95% confidence interval [CI], -4.00% to -3.13%) annually. The average annual percentage change (AAPC) between 2000 and 2013 was higher in Hispanics (AAPC, -5.72%; 95% CI, -7.19% to -4.23%) than in non-Hispanic whites (AAPC, -4.53%; 95% CI, -5.32% to -3.74%) and in Blacks (AAPC, -3.70%; 95% CI, -4.34% to -3.06%). While KS incidence decreased by 5.39% per year among 30-44-year-old individuals (95% CI, -7.19% to -4.23%), the incidence of KS increased among 20-29-year-old individuals (AAPC, 2.48; 95% CI, 0.83% to 4.16%) and remained stable among 45-54-year-old individuals (AAPC, -0.26; 95% CI, -0.95% to 0.44%). Between 2003 and 2013, KS incidence rates were highest in Georgia (2.69/100,000), New York (2.07/100,000), California (1.91/100,000), Florida (1.81/100,000) and Texas (1.32/100,000). While overall incidence decreased in New York (AAPC, -4.64%; 95% CI, -6.65% to -2.45%), California (AAPC, -4.95%; 95% CI, -6.65% to -3.21%), Florida (AAPC, -7.56%; 95% CI -9.43% to -5.65%) and Texas (AAPC, -4.43%; 95% CI -7.29% to -1.49%) between 2003 and 2013, incidence rates remained stable in Georgia (AAPC, 2.35%; 95% CI -1.14% to 5.96%). Conclusion: Geographic and racial disparities in KS incidence remain. Georgia has the highest incidence rates, and, unlike in other high incidence states where incidence has decreased, the incidence of KS has remained unchanged in Georgia. 655 KNOWLEDGE OF KAPOSI SARCOMA (KS) AND HIV AMONG TRADITIONAL HEALERS IN ZIMBABWE Suzanne P. Fiorillo 1 , Francis Jaji 2 , Maxwell Matimba 2 , Thomas Campbell 1 , Margaret Borok 2 1 University of Colorado Anschutz Medical Campus, Aurora, CO, USA, 2 University of Zimbabwe, Harare, Zimbabwe Background: Traditional healers play a major role in providing health care in Africa and many HIV-infected people seek care from traditional healers while simultaneously receiving Western-style medical care. This pluralistic practice can result in treatment delays, late diagnosis and higher mortality rates for HIV- related conditions, including Kaposi sarcoma (KS). This study aimed to evaluate traditional healer knowledge of HIV and KS and the perceptions of people with KS who have sought care from traditional healers. Methods: Between September 2015 and February 2016, traditional healers were invited to attend one of 11 informational sessions conducted by KS medical experts at 8 clinics in Zimbabwe. Healers were asked to complete a survey about their knowledge and experience in treating KS and HIV. KS clients were asked to complete a survey about their perceptions towards traditional healers. Qualitative results were organized into themes and analyzed. Results: Data were obtained from 406 traditional (30%), faith (32%) and herbalist (38%) healers. Faith and herbal healers were more likely to be older and more educated than traditional healers (both p<0.0001). Among healers, 76% claimed familiarity with KS and had treated clients with the disease. Over half (55%) of healers had positive opinions toward medical treatment only for KS and HIV, and 20% of healers recommended that both traditional and medical practitioners work together in treating the diseases. Healers were more knowledgeable of the accepted medical treatment for HIV but had limited knowledge of treatment options for KS. Healers would more often recommend traditional medicines; such as herbs, supplements, dietary concoctions or spiritual guidance, for treating KS than for HIV. Surveys from 395 KS clients were obtained, with 115 (29%) having visited a healer before. Those who had visited a healer were more likely to be rural rather than urban dwellers (OR 1.6; p=0.05) and 82% visited a healer specifically for KS consultation; 17%were ‘somewhat’, ‘very’ or ‘definitely’ satisfied with the care received from a healer whereas 94% had the same satisfaction with the care received at the medical clinic. Conclusion: Traditional healers are often the first care provider for many seeking KS treatment and a collaborative approach toward care should be considered. Even though knowledge of medical treatment of HIV has increased, better understanding of KS disease by traditional healers could help with earlier diagnosis and linkage to medical care.

653 RISK OF NON-AIDS-DEFINING CANCERS AMONG VETERANS WITH WELL- CONTROLLED HIV INFECTION Aaron P. Thrift , Kathryn E. Royse, Peter A. Richardson, Suchismita Raychaudhury, Roxanne Desiderio, Donna L. White, Jennifer R. Kramer, Elizabeth Chiao Baylor College of Medicine, Houston, TX, USA Background: The introduction of combined antiretroviral therapy has revolutionized HIV-infection management, resulting in improved outcomes and survival for HIV-infected individuals. However, as individuals with HIV are living longer and aging, their risk of mortality from cancer has increased. We aimed to examine risk of non-AIDS-defining cancers (NADC) in a large contemporary cohort of patients with well-controlled HIV infection during the antiretroviral therapy era. Methods: This was a retrospective cohort study from a total of 121 facilities in the Veterans Health Administration. Veteran patients with HIV diagnosis between 1 October 1999 and 31 December 2016 were included and followed until cancer diagnosis, death or 12/31/2016. NADCs were identified using the VA Clinical Cancer Registry, and included esophageal, stomach, lung, head and neck, colorectal, prostate, liver, anal, oropharyngeal, and Hodgkin lymphoma. We compared the incidence rate of all NADCs combined in veterans with HIV with those in a matched cohort (4:1 matched on age, sex, and date of HIV- matched diagnosis) of veterans without HIV infection. We also report incidence rates separately for esophageal, stomach, lung, and prostate cancers. Results: We identified 46,765 patients with HIV infection who met our study eligibility criteria. Most were men (96.9% of follow-up time) and aged 40 to 59 years at HIV infection (65.7%). Further, African Americans (48.2%) and whites (40.9%) were the two largest race/ethnicity groups. During 430,595 person- years of follow-up, 4020 patients develop a NADC (all cancers combined), yielding an incidence rate of 9.34 per 1,000 person-years (95% confidence interval [CI] 9.05-9.63). Incidence rates were highest among persons aged >70 years and Asians. The incidence rate of all NADCs combined was almost 5-fold higher among the HIV cohort relative to the non-HIV cohort (incidence rate, 1.96 per 1,000 person-years). Among HIV-infected patients, risk of esophageal, stomach, lung and prostate cancer were 2.93-fold (0.17 vs. 0.06 per 1,000 person-years), 2.86-fold (0.12 vs. 0.04 per 1,000 person-years), 4.64-fold (2.22 vs. 0.48 per 1,000 person-years), and 3.52-fold (2.71 vs. 0.75 per 1,000 person- years) higher relative to the non-HIV cohort, respectively. Conclusion: People with HIV infection are at increased risk for developing common NADCs compared to age-matched controls in the antiretroviral therapy era. Further research is needed to understand the reasons for this increased risk. 654 TRENDS IN INCIDENCE OF KAPOSI SARCOMA AMONG MALES IN ALL 50 UNITED STATES, 2000-2013 Aaron P. Thrift , Donna L. White, Harrison Nguyen, Kathryn E. Royse, Jennifer R. Kramer, Elizabeth Chiao Baylor College of Medicine, Houston, TX, USA Background: Kaposi sarcoma (KS) is the most common neoplasm of people living with HIV today. Although the overall incidence of KS has been reported to be declining in the US, KS has strong racial/ethnic, age, and regional diversity in incidence trends. Methods: We analyzed KS incidence data from the US Cancer Statistics (USCS) registry for the years 2000-2013. The USCS registry is the official data source for federal government–reported cancer incidence statistics and covers 97% of the US population. Women were excluded because of the low numbers of KS cases in certain geographic regions. We also restricted our analyses to the 20 to 54

Poster Abstracts

CROI 2018 243