The Pattern of Cervical Cancer according to HIV Status in Yaoundé, Cameroon

Objective To analyze the epidemiological aspects of invasive cervical cancer according to HIV status. Methods This was an historical cohort study from January 2010 to April 2017 in three hospitals at the Yaoundé city Capital, Cameroon, after the National Ethics Committee' approval. We included invasive cervical cancers with documented HIV status. Odds ratios and 95% confidence interval were calculated to assess the association between the different variables and HIV status. Survival was analyzed using the Kaplan–Meier. The level of significance was set up at <5%. Results Among the overall 213 cervical cancer patients, 56 were HIV+ (24.67%). Factors associated with positive HIV status were age below 40 (OR: 2.03 (1.38–2.67)), celibacy (OR: 2.88 (1.58–4.17)), nonmenopausal status (OR: 2.56 (1.36–3.75)), low parity, primiparity (OR: 2.59 (1.43–3.74)), and for parity with 2–4 children (OR: 2.24 (1.35–3.12)). Concerning the HIV+ patients, tumor was diagnosed late (stages III-IV) (OR: 2.70 (1.43–5.08)), undifferentiated (grade III) (OR: 7.69 (5.80–9.57)), with low median survival (9.83 months vs. 20.10 months). Conclusion HIV is frequent among cervical cancer patients. In the HIV+ patients, the diagnosis was made at the advanced stage, cells were poorly differentiated, and the prognosis was worse.


Introduction
Cervical cancer is a malignant neoformation developed at the opening of the cervix, most often on the exocervical mucosa [1]. Cervical cancer starts by a precancerous lesion known as benign and asymptomatic epithelial abnormality. Proper management of these precancerous lesions prevents development towards invasive cancer [1,2].
In resource-limited countries, cervical cancer is the second leading cause of death from cancer in women, after breast cancer [1].
is sexually transmitted disease progresses very slowly between 10 and 20 years from infection to premalignant and malignant lesion giving a great possibility of screening [2]. In 2018, GLOBOCAN estimated that there were more than 570,000 new cases diagnosed a year, and 311,000 women die during the same period [3]. GLOBO-CAN also estimated that there were more than 32,000 new cases of cervical cancer in Central Africa, with about 23,000 deaths during the same year [3].
ere is a wide geographical disparity in the distribution of cervical cancer worldwide, with over 85% of cases in developing countries [3].
is disparity in geographical distribution of cervical cancer, with high morbidity and mortality in developing countries, is mainly due to the absence of systematic screening for this cancer and comorbidity with the advent of HIV/AIDS pandemic in these poor regions [3,4].
HIV infection is also a real public health issue worldwide and mainly in African countries leading to critical consequences, including morbidity and mortality [4]. HIV infection generally increases the risk of cancer and mainly the viral-induced one [4]. Many studies documented that HIV infection and the subsequent immunodeficiency are associated with increased risk of precancerous lesions [4][5][6].
erefore, HIV-infected women are 4-9 times more likely to develop cervical cancer [7,8]. In 1993, the revised classification by CDC (Centre of Disease's Control and Prevention), included the invasive carcinoma of the cervix among the diseases that move from HIV's infection stage to AIDS disease one [9].
In Cameroon, cervical cancer is the second cancer in women, just after breast cancer (25.2% vs. 35.1%) [3]. Moreover, in some series, cervical cancer stands before breast cancer [10,11]. In the same country, HIV has a prevalence of 1.9% in men aged 15-64 and 3.6% in women of the same age group [12].
Little is known about the coexistence of the two diseases in Cameroon.

Objective.
To analyze the epidemiological aspects of invasive cervical cancer according to HIV status in Yaoundé, Cameroon.

Type-Place-Period.
is is an historical cohort study over a 7-year period (January 2010-April 2017) in Yaoundé, Cameroon. e three study sites were Yaoundé University Hospital Centre (YUTH), Yaoundé General Hospital (YGH), and Yaounde Gynaeco-Obstetric and Paediatric Hospital (YGOPH). e study was approved by these hospitals.

Study Population.
e study population included women in whom invasive cervical cancer was detected. After obtaining their verbal consent, all patients with invasive cervical cancer with documented histological diagnosis and HIV status were included.

Variables.
Data on HIV status, sociodemographic, reproductive, clinical, and therapeutic characteristics of patients were collected. Data collection was done from medical records and phone calls after verbal consent.

Data Analysis.
e data were analyzed using Epi-Info 7.2.2.6 software (CDC, USA) and SPSS 20.0 software (IBM, Armonk, NY, USA). e absolute and relative frequencies as well as central tendency (mean and median) and dispersions parameters were calculated. e odds ratios were calculated within 95% confidence interval in order to assess the influence between the different variables and HIV status. e dependent variables considered were the HIV status of patients with invasive cervical cancer and the survival period. Pearson, Fisher, and Wald statistical tests were used. e level of significance was set up at p < 5%.

Results
A total of 213 invasive cervical cancer patients were recorded, including 56 with HIV+ status (24.67%). e median age of invasive cervical cancer patients was 12 years lower for HIV+ patients compared to their HIV− status counterparts ((43 vs. 55 years); p < 0.001).

Discussion
We found an HIV frequency of 24.67% in cases of invasive cervical cancer; therefore, about 1 out of 4 women was HIV+. In Cameroon, the average frequency of HIV infection (reported by the 2018 Demographic Health Survey) is 2.9%, with some gender disparities: 1.9% in men and 3.6% in women [12]. e high frequency of HIV-positive status among patients with invasive cervical cancer compared to HIV prevalence in the general population in women may be due to the fact that HIV infection increases the risk of   [13,14]. Others found proportions higher than ours (29.7-66.4%) [15][16][17][18][19][20]. A study in Botswana reported a prevalence of 66.4% [18]. ese differences may be due to the fact that HIV prevalence in women in Cameroon (3.6% according to the latest Demography Health Survey) is lower than HIV prevalence in Botswana (23.4% in 2013, according to the UNAIDS report) [12,21]. As for the factors associated with HIV+ status, we found that, compared to older women (40-60 years), younger ones (<40 years) had twice the risk of being HIV+ (OR: 2.03 (1.38-2.67)). Our findings may be due to the fact that HIV infection accelerates the development of precancerous lesions to invasive cancer. ese findings corroborate those of many other authors related to the young age of HIV+ patients with cervical cancer at diagnosis. A study in Tanzania revealed findings similar to ours. ey found that young patients with invasive cervical cancer under 50 years were more likely to be HIV+ [20].
Single women were more likely to be HIV+ among ICC patients (OR: 2.88 (1.58-4.17)). Similarly, separated/ divorced women were more likely to have increased risk of being HIV+ (OR: 3.08 (1.69-4.46)). ese findings may be partly due to the fact that people not involved in a stable exclusive relation are much more exposed to HIV. is report on separated/divorced status is similar to that of the 2011 DHS-MICS; HIV prevalence was highest in widows (17.9%) and divorcees (15.7%) [22].
Compared to patients living in rural areas, those living in urban ones were more likely to be HIV+ (OR: 3.66; 95% CI: 1.14-6.17).
ese findings corroborate those of the 2018 HDS-MICS, which reported the higher HIV prevalence in women living in urban areas compared to those living in rural ones (3.9% vs. 2.9%) [12].
As for parity, our findings revealed that women of low parity were more likely to be HIV+. ese findings may be due to the fact that HIV+ status could decrease the desire for maternity in some women. is could also be due to the fact that ICC occurs more in HIV+ women at a younger age [8].
Women with a history of cumulated sexual partners (CSP) ≥ 5 were more likely to be HIV+, compared to those with less than 5 CSP (OR: 2.24 (1.35-3.12)). ese findings may be due to the fact that a history of many sexual partners is a common risk factor for HIV and ICC [2]. ese findings are similar to those of a study in Tanzania that reported a history of many sexual partners as a factor associated with HIV+ status (>6) (OR: 5.56; 95% CI: 1. 18-26.25) [20].
HIV+ patients mostly came with stages III-IV and grades II and III (57.15%). Other authors highlighted the association between HIV and poorly differentiated tumor cells [23]. ese findings are similar to those in Tanzania that found the possibility six times greater to be HIV+ in women with tumor poorly differentiated cells than those with good or moderate differentiated ones [24].
In HIV+ women, the median of survival was worse than in HIV− ones (9.83 months vs. 20.10 months). erefore, half of HIV+ patient with ICC died in the same year, before the tenth month; half of HIV− ones died before the second year, at least during the first 20 months. Lower survival rates in HIV+ patients with ICC compared to HIV− ones were underlined by other authors [24][25][26][27]. ese results may be  due to the advanced stage at diagnosis and poor cell differentiation in HIV+ patients [24]. HIV+ patients with ICC could die sooner of opportunistic infections probably due to the deteriorating of immune-suppression condition. Our findings are similar to those of Dryden Peterson et al., in 2016 in Botswana, who also found that HIV+ patients had a shorter median survival than HIV− ones (21.7 months vs. 30.5 months), i.e., a 9-month difference between the 2 groups [18]. Median's survival of HIV+ and HIV− found in our study are 12 months lower than those found in Botswana. is significant difference may be due to different therapeutic modalities in each of the two countries, grades disparity of tumoral differentiation and treatment compliance.

Conclusion
HIV infection prevalence is high in patients with ICC (24.67%). HIV+ patients with ICC are younger, most often received advanced stages III-IV, and have mostly poorly differentiated cells. HIV+ patients with ICC have survival median lower than HIV− ones. We emphasize on the need to create multidisciplinary care units for people living with HIV+ including cervical cancer prevention.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
Pierre Marie Tebeu initiated the study, analyzed data, supervised, and wrote the article. Laure Leka Zingué drafted the protocol and collected and analyzed data. Jesse Saint Saba Antaon analyzed data and wrote and submitted the article. Etienne Okobalemba Atenguena and Julius Sama Dohbit corrected the protocol and supervised data collection. Jean Pierre Ngou-Mve-Ngou critically reviewed the article.