Conjunctival squamous cell carcinoma in people with HIV in South Africa: time to renew efforts for novel oncogenic virus discovery?

Conjunctival squamous cell carcinoma is a rare cancer whose incidence is suspected to be causally related to exposure to solar ultraviolet light, based on findings of a strong inverse association between incidence and latitude (an increase of 49% per 10 degrees decline) (1). It is possible that this association may be mediated by ultraviolet B, which is known to damage epithelial cells (2) and which would also explain other associations of conjunctival squamous cell carcinoma [eg, with spending time outdoors and a prior history of skin cancer (3)]. Around 1989, HIV infection emerged as a new risk factor for conjunctival squamous cell carcinoma when reports of cases diagnosed in people with HIV in the United States appeared in the literature (4,5). These reports were followed shortly thereafter with case series reports of a high frequency of conjunctival neoplasms in Uganda and Rwanda in Africa (6,7), 2 countries at the epicenter of the HIV pandemic at the time (8). The cases diagnosed in people with HIV were different from those in the pre-HIV pandemic era in being diagnosed at a younger age, exhibiting rapid tumor growth, and a highly malignant histological appearance. These features suggested to the authors that HIV was associated with developing conjunctival squamous cell carcinoma and a high index of suspicion for HIV should be applied when evaluating new cases of conjunctival squamous cell carcinoma. To epidemiologists, the conjunctival squamous cell carcinoma case reports were reminiscent of those for Kaposi sarcoma (KS) (9), which heralded the outbreak of the HIV-AIDS pandemic a decade earlier and paved the way for establishing a strong link between HIV, immunosuppression, and unusual malignancies. Epidemiological and clinical studies of KS established the principles now widely used to assess and evaluate associations of cancers with HIV and the categories of AIDS-defining, non–AIDS-defining, or incidental cancer in people with HIV to guide clinical decisions and research priorities. These principles include 1) a strong temporal association between the incident malignancy and HIV; 2) a consistently (and sometimes dramatically) elevated risk of the malignancy in people with HIV compared with those without HIV; 3) a trend of increasing risk with worse levels of immunosuppression; 4) a decrease in risk when immunocompetence is restored with effective antiretroviral treatment of HIV; and often 5) demonstration or discovery of an infection, which is dysregulated by immunosuppression, thereby triggering progression to malignancy. KS met all these criteria, including the discovery of a novel oncogenic virus, KS–associated herpesvirus (also known as human herpesvirus 8) (10), which was confirmed to be causally associated with KS (11) and sufficient to explain KS epidemiology globally (12).

In contrast to KS, expectations that conjunctival squamous cell carcinoma would be confirmed as another AIDS-defining cancer and would become a tumor targeted for oncogenic virus discovery fell short. The studies conducted in regions of Africa that have a substantial HIV burden (8) and cancer registries with long-term data, such as Kyadondo in Uganda (13-15) and Harare in Zimbabwe (16), confirmed a strong association with HIV. Temporal increases in conjunctival squamous cell carcinoma incidence (6- to 13-fold) with HIV were reported (17), but variation in study designs, quality of tumor diagnosis in the African studies, and inconsistency of findings in different reports led to a more cautious interpretation of data and limited their impact on research. For example, a large study conducted in urban Black South African patients in the early 2000s (18), when the HIV epidemic was expanding rapidly in the population, failed to find an association of conjunctival squamous cell carcinoma with HIV. Two studies conducted in the United States using the HIV/AIDS Cancer Match Study (19,20) confirmed an approximate 12-fold increase in risk for conjunctival squamous cell carcinoma with AIDS, but these studies were not able to demonstrate temporal trends or associations of risk with severity of immunosuppression (20). Although these studies were large (50 000 and 490 000 people with HIV), they identified few cases (4 and 15 cases, respectively), which reduced interest in conjunctival squamous cell carcinoma by researchers in North America and Europe, particularly for virus discovery. Moreover, a few studies conducted to investigate potential associations of conjuctival squamous cell carcinoma with human papilloma viruses (HPVs), which were potentially low-hanging fruit for discovery of novel virus-cancer associations based on prior associations of skin cancer with HPV (3), yielded null (21) or conflicting results (22), further dampening enthusiasm for virus discovery research in conjunctival squamous cell carcinoma.

In the current issue, Metekoua and colleagues (23) report trends in conjunctival squamous cell carcinoma in a cohort of people with HIV in South Africa during 2004-2014 (24). The study was conducted in the South African HIV Cancer Match Study that was constructed by probabilistic record linkage of a list of people with HIV in South Africa, obtained from the National Health Laboratory Service, to the National Cancer Registry database, while preserving patient privacy and confidentiality. The National Health Laboratory Service covers 80% of the South African population, making it broadly representative of people receiving public health services in South Africa. With a sample size of 5 247 968 people with HIV and 1059 incident cases of conjunctival squamous cell carcinoma, this is the largest and most representative study of conjunctival squamous cell carcinoma conducted to date. The authors found that the incidence of conjunctival squamous cell carcinoma in South African people with HIV was 65% higher during 2004 to 2009, when there was limited antiretroviral treatment, than during 2010-2014, when antiretroviral treatment was more available. These temporal changes may be attributed to a lower risk of conjunctival squamous cell carcinoma in people with HIV on immune-restoring antiretroviral treatment than those who are not. The authors used availability of baseline CD4 count cells, which reflect the degree of immunosuppression, to more directly assess the link between conjunctival squamous cell carcinoma and immunity. Compared with those with at least 500 CD4 cell counts at baseline, the risk of conjunctival squamous cell carcinoma was 60% higher among those with CD4 counts of 350-500 and fivefold higher among those with CD4 counts of no more than 200. Moreover, temporal analyses showed that the incidence of conjunctival squamous cell carcinoma declined by 10.9% between 2004 and 2014 (-11.8% in males and -9.6% in females), which is consistent with decreases in risk in people with HIV receiving immune-restoring antiretroviral treatment. Not surprisingly, living at lower latitudes, which is an established risk factor for conjunctival squamous cell carcinoma (1), was associated with conjunctival squamous cell carcinoma indicating that immunosuppression increases the risk of this carcinoma on the background of ultraviolet solar radiation. Although the conjunctival squamous cell carcinoma risk was similar between males and females, these data suggest that middle-aged males appear to have higher risk than middle-aged females (23).

So, what is the clinical significance of the study by Metekoua and colleagues (23)? The study provides strong unequivocal evidence that conjunctival squamous cell carcinoma is associated with HIV and fulfills the criteria for HIV-AIDS association that eluded earlier researchers (25). The associations with CD4 counts provide a direct link with immunosuppression, the temporal associations mirror those expected based on improvements in the immune status of people with HIV receiving treatment, and provide compelling evidence that reduction in risk for conjunctival squamous cell carcinoma is an additional benefit from public health measures implemented to prevent people with HIV from developing severe immunosuppression thanks to antiretroviral treatment availability and utilization.

The study sheds light on contradictory findings of conjunctival squamous cell carcinoma in South Africa—initially null with HIV (18) but recently found to be elevated 18.7-fold with HIV during 1995-2016 (26). In this recent report, the risk of conjunctival squamous cell carcinoma, measured as odds ratio of association of HIV with conjunctival squamous cell carcinoma, rose from 7.6 in the pre–antiretroviral treatment era (1995-2004) to 35.7 in the later antiretroviral treatment era during 2010-2016, suggesting that the risk of conjunctival squamous cell carcinoma may actually be rising. We suspect that such findings probably reflect, at least in part, ascertainment and survivorship biases—with a deficit of conjunctival squamous cell carcinoma diagnoses in earlier years because of a paucity of antiretroviral treatment and other clinical care and a surfeit of diagnoses with increasing antiretroviral treatment and better care.

Metekoua and colleagues’ work (23) provides benchmarks for future conjunctival squamous cell carcinoma work and public health policy, given that they address the criteria to establish associations with HIV-AIDS–associated malignancies mentioned above. However, the study falls short on the possibility of viral infection(s) in conjunctival squamous cell carcinoma etiology. This unanswered question could prove to be a fertile area for research to build on this important work. Metekoua and colleagues (23) confirm that latitude is an important cofactor in the development of conjunctival squamous cell carcinoma, presumably as a proxy for exposure solar ultraviolet radiation and associated DNA damage that predisposes to conjunctival squamous cell carcinoma development. A better understanding of this interaction may inform conjunctival squamous cell carcinoma research, clinical care, and prevention strategies in collaboration with local resources, such as South Africa’s National Health Laboratory Service. These findings may encourage investigators in the tropical belts, where both conjunctival squamous cell carcinoma and HIV co-occur, such as India (27), to conduct similar studies or collaborate with South Africans to discover the cofactor upregulated by immunosuppression to cause conjunctival squamous cell carcinoma. Finally, the current study raises cost-benefit questions regarding interventions to prevent conjunctival squamous cell carcinoma in people with HIV. Although the costs of antiretroviral treatment are easily justified based on other considerations, more information on the additional costs of screening eyes, biopsy, and counseling would clarify the value of prevention, early detection, and/or treatment of conjunctival squamous cell carcinoma in groups at increased risk.

Data availability

No new data were generated or analyzed for this editorial.

Author contributions

Jennifer K. McGee-Avila, PhD (Writing—original draft) and Sam M. Mbulaiteye, MBChB (Conceptualization; Supervision; Writing—review & editing).

Funding

No funding was uses for this editorial. The authors are supported by the Intramural Research Program of the Division of Cancer Epidemiology and Genetics, National Cancer Institute.

Conflicts of interest

Jennifer K. McGee-Avila and Sam M. Mbulaiteye have no disclosures.