Age patterns of Kaposi's sarcoma incidence in a cohort of <scp>HIV</scp>‐infected men

<jats:title>Abstract</jats:title><jats:p>The life expectancy for <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐positive individuals has improved over time due to increasing access to highly active antiretroviral therapy (<jats:styled-content style="fixed-case">HAART</jats:styled-content>). Yet, as the <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐positive population ages, their risk of developing cancers also increases. Studies of Kaposi's sarcoma (<jats:styled-content style="fixed-case">KS</jats:styled-content>) among elderly <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐infected persons are quite limited. We examined the age patterns of <jats:styled-content style="fixed-case">KS</jats:styled-content> incidence and an association between age and <jats:styled-content style="fixed-case">KS</jats:styled-content> risk in a <jats:styled-content style="fixed-case">US</jats:styled-content> cohort of 3458 <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐infected men, the Multicenter <jats:styled-content style="fixed-case">AIDS</jats:styled-content> Cohort Study (<jats:styled-content style="fixed-case">MACS</jats:styled-content>). Poisson distribution was used to calculate incidence rates and respective 95% confidence intervals (95% <jats:styled-content style="fixed-case">CI</jats:styled-content>s). Cox proportional hazards regression was performed to examine the association between age and <jats:styled-content style="fixed-case">KS</jats:styled-content> risk. There were 534 incident <jats:styled-content style="fixed-case">KS</jats:styled-content> cases with a total follow‐up time of 25,134 person‐years. The overall <jats:styled-content style="fixed-case">KS</jats:styled-content> incidence rate was 2.13 per 100 person‐years (95% <jats:styled-content style="fixed-case">CI</jats:styled-content>: 1.95–2.32) (Non‐<jats:styled-content style="fixed-case">HAART</jats:styled-content> users‐ever: 5.57 per 100 person‐years [95% <jats:styled-content style="fixed-case">CI</jats:styled-content>: 5.09–6.10]; <jats:styled-content style="fixed-case">HAART</jats:styled-content> users‐ever: 0.39 per 100 person‐years [95% <jats:styled-content style="fixed-case">CI</jats:styled-content>: 0.31–0.51]). Overall, <jats:styled-content style="fixed-case">KS</jats:styled-content> frequency and incidence declined with age, even in the oldest age group (<jats:italic>p</jats:italic><jats:sub>trend</jats:sub> < 0.0001). However, among non‐<jats:styled-content style="fixed-case">HAART</jats:styled-content> users‐ever, the oldest age group had the highest incidence rate ratio compared to younger individuals [15.01, 95% <jats:styled-content style="fixed-case">CI</jats:styled-content>: 6.12–44.22]). While the incidence of <jats:styled-content style="fixed-case">KS</jats:styled-content> decreased with age, older <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐infected persons who do not receive <jats:styled-content style="fixed-case">HAART</jats:styled-content> are still at increased risk of <jats:styled-content style="fixed-case">KS</jats:styled-content>. As <jats:styled-content style="fixed-case">KS</jats:styled-content> remains an important malignancy among <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐infected persons, earlier <jats:styled-content style="fixed-case">HIV</jats:styled-content> diagnoses and <jats:styled-content style="fixed-case">HAART</jats:styled-content> initiation, particularly in older <jats:styled-content style="fixed-case">HIV</jats:styled-content>‐infected persons is warranted.</jats:p>