The impact of the South African antiretroviral treatment programme on the age-standardised incidence rate of Kaposi sarcoma, 1999-2016: An interrupted time series analysis

Artificial Intelligence-based Diagnosis of Kaposi Sarcoma using Photographs in Dark-skinned Patients

Importance

Advanced-stage disease at the time of diagnosis, with resultant high mortality, is among the most urgent issues for HIV-related Kaposi sarcoma (KS) in sub-Saharan Africa. Lack of access to skilled clinical personnel and histopathologic technology in the region contribute to diagnostic delays and advanced stage at diagnosis. Accordingly, new paradigms for KS diagnosis are needed.

Objective

To evaluate the accuracy of artificial intelligence (AI)-based interpretation of digital surface images of skin lesions to diagnose KS among dark-skinned patients in Uganda.

Design

Cross-sectional study of consecutive participants referred to skin biopsy services in Uganda because of clinical suspicion of KS. Lesions were photographed using a digital camera, and punch biopsies were obtained. Histopathologic interpretation was considered the gold standard. Using training (∼70% of images) and validation (∼10% of images) sets, we developed a prediction model using a rule-based combination of YOLO (You Only Look Once) version 5 and 8 object detection classifiers.

Setting

Free-of-charge skin biopsy services.

Participants

Consecutive sample of 482 individuals were evaluated due to clinical suspicion of KS.

Main Outcomes

Sensitivity, specificity, positive and negative predictive value (with accompanying 95% confidence intervals) of the AI-based prediction model in a test set (∼20% of images). The accuracy of a dermatologist's visual interpretation of images was also described.

Results

472 participants (1385 images) were evaluable. Of these, 36% were female, median age was 34 years, and 94% had HIV; 332 had KS, and 140 had no KS by histopathology. In the test set, the AI-derived prediction model achieved 89% sensitivity (85%-94%) and 51% specificity (40%-61%) for diagnosing KS; positive predictive value was 81% (75%-86%) and negative predictive value was 67% (55%-78%). A dermatologist evaluating the same images, with emphasis on sensitivity, achieved sensitivity of 93% (89%-96%) and specificity of 19% (11%-28%).

Conclusions and Relevance

Among dark-skinned patients in Uganda with skin lesions suspicious for KS, evaluation of digital surface images by an AI-based prediction model produced moderate accuracy for diagnosing KS. While currently inadequate for clinical use, this inaugural assessment is sufficiently promising to justify evaluation of larger datasets and evolving technologies to determine if accuracy can be improved.

Key Points

Question: Can an artificial intelligence (AI)-based prediction model be developed from digital images to accurately distinguish Kaposi sarcoma (KS) from non-KS in dark-skinned patients?Findings: Evaluation of digital images of skin lesions from patients in Uganda by an AI-based prediction model produced moderate accuracy for diagnosing KS.Meaning: In sub-Saharan Africa, where incidence and mortality from KS is high and delayed diagnosis is common due to limited specialized personnel and technical supplies, AI-based prediction models built on digital images taken of suspicious lesions may someday hasten KS diagnoses.

Patterns of incident Burkitt lymphoma during the HIV epidemic among the Black African and White population in South Africa

BACKGROUND

Burkitt lymphoma (BL) may be HIV-associated but data on BL trends in South Africa (SA), where HIV is highly prevalent, are scarce. We compared BL incidence trends over 36 years among Black African and White individuals.

METHODS

We included histologically diagnosed BL from the National Cancer Registry in SA between 1986-2021. We computed yearly age-standardised incidence rates (ASIR) by race, and annual percentage changes in ASIR using Joinpoint regression.

RESULTS

Between 1986-2021, 2205 Black African (ASIR: 1.68/1,000,000; 95% confidence interval [CI] 1.63-1.73) and 366 White individuals (ASIR: 2.34/1,000,000; 95% CI 2.15-2.53) had incident BL. Median age at diagnosis increased over time, while the male proportion among those diagnosed declined. The ASIR among Black Africans increased from 1986-2012 and declined thereafter with BL incidence peaks shifting from children and elderly to middle-aged adults. Among White individuals, BL rates rose among all age groups over time.

CONCLUSIONS

The BL epidemiology among Black Africans, with decreasing rates since 2012, may reflect SA's evolving HIV epidemic. In contrast, BL rates among White individuals in SA and many high-income countries continue to increase over time. Further studies are needed to better understand the differences in BL epidemiology across geographic regions and population groups.

Survival Following Diagnosis of HIV-Associated Kaposi Sarcoma Among Adults in East Africa in the "Treat-All" Era

Background

Despite widespread access to antiretroviral therapy (ART) in the "Treat All" era, HIV-associated Kaposi sarcoma (KS) remains among the most common malignancies in sub-Saharan Africa. Survival after KS diagnosis has historically been poor in Africa, but knowledge whether survival has changed at the population level in the contemporary era has been limited by lack of community-representative surveillance and monitoring systems.

Methods

We identified all adult persons living with HIV (PLWH) with a new diagnosis of KS made between 2016 and 2019 during outpatient or inpatient care at prototypical primary care-providing medical facilities in Kenya and Uganda using rapid case ascertainment. Participants were subsequently followed for vital status, including community tracking for those who became lost to follow-up.

Findings

Among 411 participants with newly diagnosed KS, 71% were men, median age was 34 (IQR: 30 to 41) years, and 91% had ACTG T1 tumor extent. Over a median follow-up of 7.8 (IQR: 2.4 to 17.9) months, cumulative incidence of death (95% CI) at months 6, 12 and 18 were 34% (30% to 39%), 41% (36% to 46%) and 45% (40% to 51%), respectively. Having the highest number of anatomic sites (11 to 16) harboring KS lesions (hazard ratio 2.2 (95% CI: 1.3-3.8) compared to 1 to 3 sites) and presence of oral KS lesions (hazard ratio 2.2 (95% CI: 1.4-3.3)) were independently associated with higher mortality. Lower hemoglobin and CD4 count as well as higher plasma HIV RNA were also associated with higher mortality.

Interpretation

Among PLWH with newly diagnosed KS in East Africa in the "Treat All" era, survival was poor and related to mucocutaneous extent of KS. The findings emphasize the need for better control of KS in Africa, including novel approaches for earlier detection, better linkage to oncologic care, and more potent therapy.

Incidence and epidemiology of conjunctival squamous cell carcinoma in relation to the HIV epidemic in South Africa: a 25-year analysis of the National Cancer Registry (1994-2018)

AIMS

To describe the incidence and epidemiology of conjunctival squamous cell carcinoma (CSCC) in South Africa over a 25-year period (1994-2018), with particular reference to the HIV epidemic.

METHODS

Incident cases of histologically diagnosed CSCC were identified from the pathology-based South African National Cancer Registry. Crude and direct age-standardised incidence rates (ASIRs) per 100 000 persons (Segi World Standard Population) were calculated using national population statistics and compared by age, sex and ethnicity. Trends in the incidence and demographic features of CSCC were described and analysed. Incidence rates were compared with national HIV-related statistics for the same time period.

RESULTS

In total, there were 9016 reported CSCC cases (women: 56.6%, black: 86.8%, mean age: 41.5 years). The overall ASIR was 0.78 per 100 000. Two distinct epidemiological patterns were identified: (1) older white men, and (2) younger black women. There was a sixfold increase in CSCC incidence rates between 1994 and 2009 with a corresponding shift from the first to the second disease profile. Despite rising HIV seroprevalence, CSCC incidence rates have declined since 2009. A strong ecological correlation (r=0.96) between CSCC incidence and widespread antiretroviral therapy (ART) provision was identified.

CONCLUSION

This study highlights the evolving trends and disease burden of CSCC in South Africa. Widespread ART provision is ecologically correlated with declining CSCC rates over the last decade. These findings are in keeping with reported trends for other HIV-related cancers and have important implications for future incidence studies and public health policy.

Prognostic analysis of cutaneous Kaposi sarcoma based on a competing risk model

The data regarding the prognosis of cutaneous Kaposi sarcoma (KS) was limited. The current study aimed to explore the risk factors and develop a predictive model for the prognosis of cutaneous KS patients. Data were extracted from Surveillance, Epidemiology, and End Results database from 2000 to 2018 and randomly divided into training and validation cohort. The Kaplan-Meier analysis, cumulative incidence function based on the competing risk model and Fine-Gray multivariable regression model was used to identify the prognostic factors and then construct a 5-, 10-, and 15-year KS-specific death (KSSD) nomogram for patients. The concordance index (C-index), area under the curve (AUC) of operating characteristics and calibration plots were used to evaluate the performance of the model. The clinical utility of the model was measured by decision curve analysis (DCA). In 2257 cutaneous KS patients identified from database, the overall median survival time was about 13 years. Radiotherapy (p = 0.013) and surgery (p < 0.001) could lower the KSSD, while chemotherapy (p = 0.042) and surgery (p < 0.001) could increase the overall survival (OS) of patients with metastatic and localized lesions, respectively. Race, number of lesions, surgery, extent of disease, year of diagnosis and age were identified as risk factors associated with cutaneous KS-specific survival. Performance of the nomogram was validated by calibration and discrimination, with C-index values of 0.709 and AUC for 5-, 10-, and 15-year-KSSD of 0.739, 0.728 and 0.725 respectively. DCA indicated that the nomogram had good net benefits in clinical scenarios. Using a competing-risk model, this study firstly identified the prognostic factors, and constructed a validated nomogram to provide individualized assessment and reliable prognostic prediction for cutaneous KS patients.

Biomedical bargains: Negotiating "safe sex" on antiretroviral treatment in rural South Africa

Wide-scale availability of antiretroviral treatment (ART) has transformed the global landscape for HIV prevention, shifting emphasis away from a strictly behavioral focus on changing sexual practices towards a biomedical approach. Successful ART management is measured by an undetectable viral load, which helps maintain overall health and prevent onward viral transmission. The latter utility of ART, however, must be understood in the context of its implementation. In South Africa, ART has become easily accessible - yet ART knowledge spreads unevenly, while counseling advice and normative expectations and experiences of gender and aging interact to inform sexual practices. As ART enters the sexual lives of middle-aged and older people living with HIV (MOPLH), a population growing rapidly, how has it informed sexual decisions and negotiations? Drawing on in-depth interviews with MOPLH on ART, corroborated with focus group discussions and national ART-related policies and guidelines, we find that for MOPLH, sexual decisions increasingly feature compliance with biomedical directives and concern for ART efficacy. Seeking consensus regarding the biological risks of sex on ART becomes an important feature of sexual negotiations, and anticipated disagreements can pre-empt sexual relationships altogether. We introduce the concept of biomedical bargains to explain what happens when disagreements arise, and the terms of sex are negotiated using competing interpretations of biomedical information. For both men and women, ostensibly gender-neutral biomedical discourses provide new discursive resources and strategies for sexual decisions and negotiations, yet biomedical bargains are still embedded in gender dynamics-women invoke the dangers of jeopardizing treatment efficacy and longevity to insist on condoms or justify abstinence, while men utilize biomedical arguments in an effort to render condomless sex safe. While the full therapeutic benefits of ART are critical for the efficacy and equity of HIV programs, they will nonetheless always affect, and be affected by, social life.

Kaposi sarcoma-associated herpesvirus, HIV-1 and Kaposi sarcoma risk in black South Africans diagnosed with cancer during antiretroviral treatment rollout

Kaposi sarcoma-associated herpesvirus (KSHV) causes Kaposi sarcoma (KS). The risk of KS is amplified in HIV-immunosuppressed individuals and antiretroviral therapy (ART) reduces KS incidence. Reliable data on the relationship between these factors are lacking in Africa. We used questionnaires and serum from 7886 black South Africans (18-74 years) with incident cancer, recruited between 1995 and 2016. ART rollout started in 2004. We measured associations between KS, HIV-1 and KSHV before and after ART rollout. We measured seropositivity to HIV-1, KSHV latency-associated nuclear antigen (LANA) and glycoprotein (K8.1) and calculated case-control-adjusted odds ratios (ORadj ) and 95% confidence intervals (CI) in relation to KS and KSHV infection, before (1995-2004), early (2005-2009) and late (2010-2016) ART rollout periods. KSHV seropositivity among 1237 KS cases was 98%. Among 6649 controls, KSHV seropositivity was higher in males (ORadj  = 1.4 [95%CI 1.23-1.52]), in persons with HIV, (ORadj  = 4.2 [95%CI 3.74-4.73]) and lower in high school leavers (ORadj  = 0.7 [95%CI 0.59-0.83]). KSHV seropositivity declined over the three ART rollout periods (37%, 28% and 28%, Ptrend  < .001) coinciding with increases in high school leavers over the same periods (46%, 58% and 67%, Ptrend  < .001). HIV-1 seroprevalence increased from 10% in the pre-ART period to 22% in the late ART period (Ptrend  < .001). Compared to HIV-1 and KSHV seronegatives, KSHV seropositives yielded an OR for KS of 26 (95%CI 11-62) in HIV-1 seronegative participants and an OR of 2501 (95%CI 1083-5776) in HIV-1 seropositive participants. HIV-1 increases the risk of KS in those infected with KSHV by 100-fold. Declines in KSHV seroprevalence coincide with ART rollout and with improvements in educational standards and general hygiene.

Thirteen cancers associated with HIV infection in a Black South African cancer patient population (1995-2016)

South Africa's HIV epidemic has evolved over time in terms of numbers of people living with HIV, access to antiretroviral treatment (ART) and age. These changes have profoundly influenced local cancer patterns. The Johannesburg Cancer Study has, over a period of 22 years (1995-2016), recruited over 20 000 incident black cancer patients who consented to provide answers to a questionnaire and blood samples (serum, DNA). This has presented a unique opportunity to examine the evolving association of HIV with cancer in Africa. We used logistic regression models to explore case-control associations between specific cancers and HIV, using participants with non-infection related cancers as controls. Using data of 20 835 cancer patients with confirmed HIV status, we found the following cancers to be associated with HIV: Kaposi's sarcoma (ORadj ; 95%CI): (99.1;72.6-135.1), non-Hodgkin lymphoma (11.3;9.3-13.6), cervical cancer (2.7;2.4-3.0), Hodgkin lymphoma (3.1;2.4-4.2), cancer of the eye/conjunctiva (18.7;10.1-34.7), anogenital cancers (anus [2.1;1.4-3.2], penis [5.4;2.7-10.5], vulva [4.8;3.5-6.4], vagina [5.5;3.0-10.2]), oropharyngeal cancer (1.6;1.3-1.9), squamous cell carcinoma of the skin (3.5;2.4-4.9), melanoma (2.0;1.2-3.5) and cancer of the larynx (1.7;1.3-2.4). Kaposi's sarcoma odds ratios increased from the pre-ART (1995-2004) to the early ART (2005-2009) period but declined in the late ART (2010-2016) period. Odds ratios for cancers of the eye/conjunctiva, cervix, penis and vulva continued to increase in recent ART periods. Our study confirms the spectrum of HIV-associated cancers found in other African settings. The odds ratios of conjunctival and HPV-related cancers continue to rise in the ART era as the HIV positive population ages.

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.

Skin Cancer in People of Color: A Systematic Review

BACKGROUND

People of African, Asian, Hispanic or Latino, Pacific Islander, and Native Indian descent are considered people of color by the Skin of Color Society (SOCS).

OBJECTIVES

In this study, we assess incidence, risk factors, clinical characteristics, histopathology, treatment, and survival for skin malignancies in people of color as defined by the SOCS, by systematically reviewing the literature.

METHODS

An electronic literature search of the PubMed, EMBASE, and MEDLINE databases was performed. Articles published from 1 January 1990 through 12 December 2020 were included in the search.

RESULTS

We identified 2666 publications potentially meeting the study criteria. Titles and abstracts of these studies were reviewed and 2353 were excluded. The full text of 313 articles were evaluated and 251 were included in this review.

CONCLUSION

Differences in incidence, patterns, treatment, and survival exist among people of color for cutaneous malignancies. Further research and initiatives are needed to account for and mitigate these differences.