Burden of Kaposi sarcoma according to HIV status: A systematic review and global analysis

In 2020, over 34 000 cases of Kaposi sarcoma (KS) were estimated globally, all attributable to KS herpesvirus (KSHV). Prior to the HIV epidemic, KS already existed in KSHV endemic regions, notably in sub‐Saharan Africa (SSA). The HIV epidemic has vastly increased the KS burden. We developed a methodology to provide global estimates of KS burden according to HIV status. A systematic review identified studies reporting HIV prevalence in consecutive KS series. Pooled estimates of HIV prevalence, by country or UN subregion, were used to calculate population‐attributable fraction (PAF) and these were applied to IARC's GLOBOCAN 2020 to estimate burden and incidence of HIV‐attributable and non‐HIV‐attributable KS. We identified 55 eligible studies, reporting HIV prevalence ranging from ≤5% to ≥95%. Approximately 80% of KS in SSA was estimated attributable to HIV, vs ~50% in the rest of the world. By applying PAFs to national GLOBOCAN estimates, an estimated 19 560 KS cases attributable to HIV were diagnosed in SSA in 2020 (~80% of the worldwide burden), vs 5064 cases of non‐HIV‐attributable KS (~60% of the worldwide burden). Incidence of HIV‐attributable KS was highest in Southern Africa (6.0 cases per 100 000) and Eastern Africa (3.4), which were also the world regions with highest incidence of non‐HIV‐attributable KS (0.4 and 1.0 cases per 100 000, respectively). This first systematic effort to produce a global picture of KS burden stratified by HIV status highlights the continuing important burden of HIV‐attributable KS in SSA, even in the era of combined antiretroviral therapy.

The epidemic form of KS (hereafter referred to as HIVattributable KS) signaled the emergence of the HIV/AIDS epidemic 2 and is vastly increased among HIV-infected persons with immunodeficiency. 4 KS incidence increased dramatically in the early phase of the HIV/AIDS epidemic, becoming the most commonly diagnosed cancer in many SSA countries. 5,6 The introduction of combination antiretroviral therapy (cART) subsequently led to a dramatic decline in KS incidence in people living with HIV (PLWHIV) in high-income settings. 7,8 As an example, relative to the general population, KS rates in PLWHIV in the United States decreased from 2800-fold in the pre-cART era (1991)(1992)(1993)(1994)(1995) and to 260-fold in 2009 to 2012. 9 However, decreases have not been uniform by geography and ethnicity, even within the United States. 10 Less marked declines have since been observed in SSA. 11,12 Despite this decrease, however, KS still accounts for significant morbidity and mortality in patients infected with HIV, 13 and incidence remains high in populations in SSA in the era of cART 1,11,12 where access and adherence to cART remain suboptimal.
There has been no recent or global attempt to systematically describe patterns of HIV-attributable (epidemic) vs non-HIVattributable (classic/endemic/iatrogenic) KS. 14 Thus, as part of the ongoing program to estimate global burden of cancer attributable to infectious agents at the IARC, 15 our aim was to provide the best possible global picture of KS burden stratified by HIV status, in the era of cART. This approach was based on a systematic review of data from KS series of known HIV status in order to estimate population attributable fractions (PAF), followed by the application of these PAFs to the GLOBOCAN estimates of KS burden at a country, world region and global level. 1 2 | METHODS

| Identification and selection of relevant studies
We performed a systematic review of four databases (PubMed, Embase, Web of Science and Global Index Medicus) to identify studies reporting the prevalence of HIV infection in consecutively diagnosed series of KS cases. Studies restricted to PLWHIV only were excluded. The details of the search strategy for each of the databases can be found in the Appendix S1 (Table S1). In brief, we used a WHO-validated search for "HIV" and "AIDS," combined with terms for "Kaposi sarcoma" and standard WHO terms for epidemiological studies. The search was last updated on 2 August 2021. There were no restrictions on language. Studies were screened for inclusion by two independent reviewers (AIK and GMC), with discrepancies resolved by consensus. If several studies used the same dataset, we included only the study with the greatest number of participants, to avoid double inclusion of patients. In order to be eligible, studies had to clearly state the period of KS recruitment (as a sign of representativeness) and to include KS cases from the year 1996 onwards (ie, to exclude studies entirely of the pre-cART era).

| Estimates of Kaposi sarcoma incidence for the year 2020
Population denominators, number of KS cases and age-standardized incidence rates (ASIR, cases per 100 000 person years) of KS were extracted for 155 individual countries/territories as described in GLOBOCAN 2020. 1

| Statistical analysis
The prevalence of HIV infection in KS cases (P c ) was assumed to equate with PAF, as in previous IARC estimates of the global burden of cancer attributable to infections. 16 Indeed, given the very high relative risk (RR) for KS in PLWHIV, even in the cART era, 17 World maps were created to display country-specific ASIRs, divided into five levels.
All statistical analyses were conducted using Stata software (Version 14.2) and world maps drawn using QGIS3 software.  Table S2.

| HIV prevalence in KS (or populationattributable fraction)
HIV prevalence in KS cases was highest (93.0%) in studies from Southern Africa ( Figure 2,

(B) Non-HIV-attributable KS
F I G U R E 3 Age-standardized incidence rates in 2020 of (A) HIV-attributable (epidemic) Kaposi sarcoma and (B) non-HIV-attributable (classic/ endemic/iatrogenic) Kaposi sarcoma. ASIR, age-standardized incidence rate. The designations used and the presentation of the material in this article do not imply the expression of any opinion whatsoever on the part of WHO and the IARC about the legal status of any country, territory, city or area, or of its authorities or concerning the delimitation of its frontiers or boundaries collectively contributed 83% of the worldwide KS burden, as estimated in 2020 (see Table S3).
ASIRs of HIV-attributable KS were highest in Southern Africa (6.0 per 100 000 person years), followed by Eastern Africa (3.4), Central Africa (1.5) and Western Africa (0.5) ( Table 1) tries. This is an approach we have previously applied to cancers attributable to other infections 16 and for cervical cancer attributable to HIV (for which the PAF was much lower than that for KS). 20,21 Since the onset of HIV/AIDS epidemic, the independent estimation of classic/endemic KS incidence has become more difficult. 22  There are relatively few data on KS incidence trends in SSA, but historical cancer registries in Uganda, Zimbabwe and South Africa have reported large increases in KS incidence in the pre-cART era of the HIV/AIDs with more moderate declines in the cART era. 11,12,23 KS incidence in the cART era has not declined as steeply in PLWHIV in SSA as it has in high-income settings 8,9,27,28  Although there have been major improvements in cancer registration in many SSA countries, 33 there also remain substantial gaps in the region where the KS burden is the highest. 34 Second, the studies identified by the systematic review included KS cases collected over a wide calendar period, for which we could only partly control for the influence of cART use on KS by excluding studies that recruited solely in the pre-cART era (ie, before 1996). This restriction was the best compromise between obtaining data relevant to PAFs in the cART era and not excluding important country-specific data. Of note, in the United States, the proportion of KS cases diagnosed among PLWHIV (ie, PAF) has been shown to remain relatively stable in the cART era, 35 despite that the risk of KS for PLWHIV has substantially declined. 9 Given that even experienced clinicians can misdiagnose KS, 18,36 histological confirmation is highly important to studies of the epidemiology of KS, and was widespread in most of the research studies contributing to PAF estimates. However, histological confirmation is lacking in a majority of KS cases in SSA, 37 and subsequently in the registry data that underpin the GLOBOCAN KS rates 11  Unfortunately, our approach did not allow attributable KS burden to be described separately by sex, nor by age. Indeed, we were not able to extract enough information from published KS series to be confident to further stratify country-specific PAFs by these variables (Table S2). KS has been strongly associated with male sex, although these sex differences are poorly understood. 2 Male/female ratios above three are still found in countries where classic KS is common or where the epidemic form has mainly involved MSM or intravenous drug users (Table S2). However, in studies from SSA, where HIV infection is widespread and predominantly transmitted through heterosexual intercourse, male/female ratio tended to be between 1.5:1 and 2:1. The ratios are consistent with GLOBOCAN 2020 KS estimates from the same countries (Table S2), and with evidence from SSA suggesting that the sex ratio has changed from 20:1 for endemic KS Infection-attributable cancer incidence estimates can raise awareness and inform primary and secondary prevention programs at a global,