CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Methods: We addressed this question using data from HIV and cancer registries in 6 states participating in the HIV/AIDS Cancer Match Study and corresponding general population mortality data from the National Center for Health Statistics (1996-2010). We compared age-stratified mortality rates in 4 groups: (1) the general population, (2) individuals diagnosed with HIV, (3) individuals diagnosed with cancer, and (4) individuals diagnosed with both HIV and cancer. We assessed associations with age-stratified overall mortality rates using a Poisson model including additive terms for HIV and cancer, and an interaction term for the combined effect of HIV and cancer. We tested the statistical significance of the interaction term to determine whether patients diagnosed with both diseases experienced significant excess mortality in addition to mortality from HIV and cancer separately. Results: For 5 of the 6 cancers evaluated, HIV-infected cancer patients experienced statistically significant excess mortality rates.( Table 1 ) This excess mortality rate was most pronounced among younger individuals (30-49 years), with large excess rates per 10 5 person-years for cancers of the lung (59,004), anus (3,736), colorectum (6,364), and breast (10,537), and melanoma (6,364); there was actually a deficit in mortality for prostate cancer in this age group (2,349 per 10 5 person-years). For breast cancer patients >70 years of age, we observed a significant excess mortality rate of 32,773 per 10 5 person-years. Conclusions: In the era of effective HIV treatment, US patients diagnosed with both HIV and cancer experienced significant excess mortality rates due to the combination of both diseases. The magnitude of this excess was substantial for non-AIDS-defining malignancies of varying etiologies, including common cancers such as lung, colorectal, and breast, and was particularly acute in younger populations who had higher baseline life expectancies.

Excess Mortality in Patients Diagnosed with both HIV and Cancer

Cancer and age

Excess Mortality Rate per 10 5 person-years

p-value

Lung Cancer

59,004

<0.0001

30-49 years

42,061

<0.0001

50-69 years

--

0.15

70+ years

Anal Cancer

3,736

<0.001

30-49 years

3,246

0.02

50-69 years

--

0.55

70+ years

Melanoma

6,364

<0.01

30-49 years

4,508

0.06

50-69 years

--

--

70+ years

Colorectal Cancer

14,952

<0.0001

30-49 years

5,856

0.08

50-69 years

--

0.66

70+ years

Breast Cancer

10,537

<0.0001

30-49 years

8,011

<0.0001

50-69 years

32,773

0.01

70+ years

Prostate Cancer

-2,349

0.03

30-49 years

--

0.15

50-69 years

--

0.35

70+ years

617 Global Burden of Cervical Cancer in HIV-PositiveWomen on Antiretroviral Therapy Eliane Rohner 1 ; Lukas Bütikofer 2 ; Mhairi Maskew 3 ;Yi-Ming A. Chen 4 ; Ruth Friedman 5 ; Gypsyamber DSouza 6 ; Matthias Egger 7 ; Julia Bohlius 1 ; for the International Epidemiologic Databases to Evaluate AIDS (IeDEA) and the Collaboration of Observational HIV Epidemiological Research in Europe (COHERE) in EuroCoord 1 Inst of Social and Preventive Med, Univ of Bern, Bern, Switzerland; 2 Univ of Bern, Bern, Switzerland; 3 Univ of the Witwatersrand, Wits Hlth Consortium, Johannesburg, South Africa; 4 Kaohsiung Med Univ, Kaohsiung, Taiwan; 5 Inst Nacional de Infectologia Evandro Chagas (INI), Fundação Oswaldo Cruz (Fiocruz), Rio de Janeiro, Brazil; 6 Johns Hopkins Bloomberg Sch of PH, Baltimore, MD, USA; 7 Univ Hosp Bern, Bern, Switzerland Background: HIV-positive women are at increased risk of human papillomavirus (HPV) infection and progression to invasive cervical cancer (ICC). HIV and HPV epidemics and access to cervical cancer screening vary between regions. We compared ICC risk in women on combination antiretroviral therapy (ART) in Asia-Pacific, North America, Latin America, Southern Africa, and Europe. Methods: We included cohorts participating in the International Epidemiologic Databases to Evaluate AIDS (IeDEA) and the Collaboration of Observational HIV Epidemiological Research Europe (COHERE) in EuroCoord. We included HIV-positive women aged ≥ 16 years who started ART after cohort enrollment from 1996 onwards. We used flexible parametric survival models with region-specific baseline hazards adjusted for time-updated CD4 cell counts, age, and year of ART start to compare regional ICC rates. We excluded Asia-Pacific frommultivariable analyses due to small sample size. We present incidence rates and adjusted hazard ratios (aHR) with 95% confidence intervals (CI). Results: We included 65,726 women from 55 countries. Median age at ART start was 35 years and similar across regions. Median CD4 cell count (cells/µL) at ART start was 115 in Southern Africa, 146 in Asia-Pacific, 179 in Latin America, and 241 in Europe and North America. Median follow-up time was 3.9 years (interquartile range 1.5-7.3). During 323,224 person-years (pys) 390 women developed ICC. Incidence rate per 100,000 pys was highest in Southern Africa (497, 95%CI 429-577) followed by Latin America (152, 95%CI 97-238), North America (76, 95%CI 48-119), Europe (71, 95%CI 62-83) and Asia-Pacific (42, 95%CI 6-297). With the exception of Southern Africa regional ICC risks decreased with time since ART start (see Figure). Adjusted hazard ratios comparing Europe with other regions at 2 and 5 years were 8.9 (95%CI 6.0-13.3) and 12.4 (95%CI 7.8-20.0), respectively, for Southern Africa, and 2.1 (95%CI 0.8-5.0) and 2.2 (95%CI 1.2-4.2), respectively, for Latin America. No difference was observed between North America and Europe. Conclusions: HIV-positive women in Southern Africa and Latin America had a markedly higher ICC risk than women from North America and Europe, and rates did not decline with time on ART in Southern Africa. These regional differences were not explained by differences in CD4 counts, age, or year of starting ART, but could be explained by a higher prevalence and incidence of HPV infection and limited access to effective cervical cancer screening.

Poster Abstracts

250

CROI 2016