Publications by Year: 2022

OBJECTIVE: The fetal risks and benefits of antihypertensives to treat gestational hypertension in pregnancy are understudied, particularly in low- and middle-income countries.

METHODS: We performed a nested case-control study within a retrospective cohort of obstetrical patients in Botswana from 2014 to 2019. We included women carrying singletons who developed new onset non-severe hypertension (140-159 mm Hg systolic or 90-109 mm Hg diastolic blood pressure) after 20 weeks of pregnancy. Cases were defined as women with either small-for-gestational-age (SGA) infants or stillbirth, analyzed separately; controls were otherwise similar women without the adverse outcome in each analysis.

RESULTS: We identified 1932 cases of SGA (7925 controls) and 316 cases of stillbirth (9619 controls). Cases with SGA were more likely to have used an anti-hypertensive than controls (33% vs 29%, adjusted odds ratio [aOR] 1.28, 95% confidence interval [CI] 1.15-1.43). Cases with stillbirth were more likely to have used an anti-hypertensive than controls (42% versus 29%, aOR 1.45, 95% CI 1.14-1.83).

CONCLUSION: Anti-hypertensive use for new-onset gestational hypertension was associated with an increased risk of having an SGA infant or a stillbirth among women who never developed severe hypertension. These data support conduct of a randomized clinical trial to determine the appropriate use of anti-hypertensives in non-severe gestational hypertension.

PURPOSE OF REVIEW: Although cervical cancer is preventable, it is the leading cancer among women in southern Africa. The association of high-risk human papillomavirus (HR-HPV) with almost all invasive cervical cancers has led to the development of effective primary and secondary prevention measures. This review focuses on updated and new evidence of the epidemiology of HPV and HPV-based secondary prevention in southern Africa.

RECENT FINDINGS: HR-HPV prevalence in southern Africa differs between regions, and varies most by HIV prevalence and age. HR-HPV prevalence among women living with HIV (WLWH) is reported between 29 and 59.7%, and between 16.2 and 25.2% among women without HIV. HPV16 is the most common HR-HPV type present in invasive cervical cancers in the region; and vaccination may potentially prevent approximately 80% of these cancers. Concerning preliminary data suggests faster development of new cervical precancer within a short follow-up period.

SUMMARY: We need tools that identify the small number of women with precancer from the many with transient HR-HPV infection in southern Africa. The high-volume of test-positive women leads to challenges in managing triage in a HR-HPV-based screening program. Longitudinal data from the entire region is urgently needed to guide effective implementation of HPV-based screening programs.

INTRODUCTION: In 2015, Botswana introduced quadrivalent human papillomavirus (HPV) vaccine for girls aged 9-13 years. To establish a baseline HPV prevalence for future HPV vaccine impact monitoring, we evaluated HPV prevalences among the youngest unvaccinated women in Botswana and compared HPV prevalences among women living with HIV (WLHIV) and without HIV.

METHODS: Women aged 18-22 years were recruited from the University of Botswana and HIV clinics in Gaborone from October 2019-January 2021. Demographic and behavioral characteristics were self-reported during structured interviews; HIV clinical characteristics were abstracted from medical charts. Self-collected vaginal swabs were tested for 28 HPV types using Seegene Anyplex II HPV28. We compared prevalence of any HPV, high risk (HR)-HPV, and quadrivalent HPV vaccine types (HPV6/11/16/18) among WLHIV and women without HIV and evaluated risk factors for prevalence of HR-HPV.

RESULTS: A total of 306 WLHIV and 500 women without HIV were recruited. Compared to women without HIV, WLHIV were more likely to be sexually experienced (86.6% versus 74.4%) and have ≥ 3 lifetime sex partners (55.3% versus 27.8%). All HPV type prevalences were significantly higher among WLHIV compared to women without HIV, including prevalence of any HPV (82.7% versus 63.0%), HR-HPV (72.9% versus 53.8%), and quadrivalent vaccine HPV types (34.3% versus 21.0%). Among WLHIV, there were no differences between those perinatally and non-perinatally infected for HPV prevalences, number of HPV types detected, CD4 count, or viral load.

CONCLUSIONS: Over one-third of WLHIV and nearly a quarter of those without HIV had vaccine-type HPV detected. This study supports need for the national HPV vaccination program in Botswana and provides important baseline data for future evaluation of impact of the program.

BACKGROUND: Women living with human immunodeficiency virus (HIV) tend to develop cervical cancer at a younger age than women without HIV. The World Health Organization's (WHO) 2021 guidelines for screening and treatment of cervical pre-cancer lesions for cervical cancer prevention include a conditional recommendation for initiating screening at age 25 for women living with HIV (WLWH). This recommendation is based on low-certainty evidence, and WHO calls for additional data. We describe the association of age and HIV status with visual inspection with acetic acid (VIA) positivity and cervical intraepithelial neoplasia grade two or higher (CIN2+) in Botswana.

METHODS: This was a retrospective cross-sectional study of 5714 participants aged 25 to 49 years who underwent VIA screening in a clinic mainly serving WLWH. VIA-positive women received cryotherapy if eligible or were referred for colposcopy and excisional treatment. Known cervical cancer risk factors, screening outcome, and histological results were extracted from the program database. We compared the proportions and association of VIA positivity and CIN2+ by age and HIV status.

RESULTS: The median age was 35 years [IQR 31-39], and 18% of the women were aged 25-29. Ninety percent were WLWH; median CD4 count was 250 cells/µL [IQR 150-428], and 34.2% were on anti-retroviral treatment (ART). VIA-positivity was associated with younger age (OR 1.48, CI 1.28, 1.72 for 25-29 years vs. 30-49 years), and HIV-positivity (OR 1.85, CI 1.51, 2.28). CIN2+ was only associated with HIV-positivity (OR 6.12, CI 3.39, 11.10), and proportions of CIN2+ were similar for both age groups in WLWH (69.1% vs. 68.3%).

CONCLUSIONS: Younger WLWH in Botswana had a significant burden of CIN2+. This finding further supports lowering the screening age for WLWH from 30 to 25.

In 2015, Botswana introduced the quadrivalent human papillomavirus (HPV) vaccine as a two-dose schedule in girls aged 9–13 years. We sought to establish a baseline HPV prevalence in unvaccinated young adults in Botswana. HIV-uninfected men and women aged 18–22 years were recruited from the University of Botswana in Gaborone during October 2019–February 2021. Demographic and behavioural characteristics were self-reported during structured interviews. Self-collected vaginal and penile swabs were tested for 28 HPV types using Seegene Anyplex II HPV28. We compared any HPV type, quadrivalent vaccine (HPV 6, 11, 16, 18)-type and non-quadrivalent vaccine-type prevalence in men and women and evaluated the risk factors for prevalence of any HPV type. A total of 493 men and 500 women were included in the analysis. Compared to men, women had higher prevalence of any HPV type (63.0% versus 31.4%, P < 0.001), vaccine-type HPV (21% versus 9.7%, P < 0.001) and non-vaccine-type HPV (60.4% versus 28.4%, P < 0.001). Higher prevalence of any HPV type in men and women was associated with having ≥2 sex partners in the past 12 months; always using condoms in the past 3 months was associated with a lower HPV prevalence. These data provide baseline information for future evaluation of the population impact of the HPV vaccination programme, including potential herd effects in men.

BACKGROUND: Delays in screening and timely diagnosis contribute significantly to global disparities in cervical cancer mortality in Botswana and other low- and middle-income countries, particularly those with high rates of HIV. Little is known about the modifiable factors shaping these delays from the perspectives of women themselves and how these perspectives may differ between those living with and without HIV.

METHODS: From March-May 2019, we conducted a concurrent, mixed methods study of women receiving treatment for cervical cancer at a multidisciplinary oncology clinic in Botswana. Enrolled participants completed a one-time, concurrent semi-structured interview and structured questionnaire assessing patient characteristics, screening and HIV-related beliefs and knowledge, and barriers and facilitators to screening and follow-up care. Qualitative data were analyzed using directed content analysis guided by the Model of Pathways to Treatment and triangulated with quantitative questionnaire data to identify areas of convergence and divergence. Fisher's exact tests were used to explore associations between questionnaire data (e.g., screening knowledge) and HIV status.

RESULTS: Forty-two women enrolled in the study, 64% of whom were living with HIV and 26% were diagnosed with stage III cervical cancer. Median age was 45 years (IQR 54-67) in those living with HIV and 64 years (IQR 42-53) in those living without. Overall screening rates before symptomatic disease were low (24%). Median time from most proximal screen to diagnosis was 52 median days (IQR 15-176), with no significant differences by HIV status. General screening knowledge was higher among those living with HIV versus those without (100% vs 73%; p < 0.05), but knowledge about HPV and other risk factors was low in both groups. Similar to questionnaire results, qualitative results indicate limited awareness of the need to be screened prior to symptoms as a central barrier to timely screening. Some participants also noted that delays in the receipt of screening results and fear also contributed to treatment delays. However, many participants also described myriad sources of social and tangible support that helped them to overcome some of these challenges.

CONCLUSION: Interventions focused on increasing routine screening and supporting timely awareness and access to care are needed to reduce global disparities in cervical cancer.

The majority of deaths from cervical cancer occur in low- and middle- income countries (LMICs). The standard of care for early-stage cervical cancer (FIGO 2018 IA2-IB1) is radical hysterectomy, a procedure performed by trained gynecologic oncologists. However, the lack of gynecologic oncologists in LMICs has required exploration into other methods of treatment for early-stage cervical cancer. A potential course of treatment for early-stage cervical cancer is neoadjuvant chemotherapy followed by simple hysterectomy and pelvic lymph node sampling, which can be performed by a general gynecologist. We gathered data for 8 women who underwent this method of treatment and found that cause-specific survival was 100% over a 3.5-year median follow-up. These findings support the exploration for this method of treatment for early-stage cervical cancer in LMICs, which would improve access to treatment for these women and hopefully reduce the high burden of cervical cancer related deaths in LMICs.

OBJECTIVE: To present the stage distribution, patterns of care, and outcomes of patients from Botswana with invasive cervical cancer, living with or without HIV.

METHODS: Between 2013 and 2020, women with cervical cancer were prospectively enrolled in an observational cohort study.

RESULTS: A total of 1,043 patients were enrolled; 69% were women living with HIV. The median age of the cohort was 47 years (interquartile range [IQR] 40-58 years), with women living with HIV presenting at a younger age compared to women without HIV (44 versus 61 years, p < 0.001). Among women living with HIV, the median CD4 count at the time of cancer diagnosis was 429.5 cells/μL (IQR 240-619.5 cells/μL), 13% had a detectable viral load, and 95% were on antiretroviral therapy. In regard to treatment, 6% (n = 58) underwent surgery, 33% (n = 341) received radiation therapy, 51% (n = 531) received chemoradiation, and 7% (n = 76) did not receive treatment. Stage distribution in the cohort was as follows: I 17% (n = 173), II 37% (n = 388), III 35% (n = 368), and IV 8% (n = 88). For all patients, 2-year OS was 67%. In multivariable Cox regression, worse OS was associated with stage: II (HR 1.91, p = 0.007), III (HR 3.99, p < 0.001), and IV (HR 5.06, p < 0.001) compared to stage I. Improved OS was associated with hemoglobin > 10 g/dL (HR 0.51, p < 0.001) compared to Hb ≤ 10 g/dL.

CONCLUSIONS: Among women in Botswana with cervical cancer, most patients presented with stage II or III disease warranting radiation therapy or chemoradiation. While two-thirds of cervical cancer patients were women living with HIV, HIV did not impact OS.