HPV-16 lineages among women in Africa: a systematic review with meta-analysis

Abstract

Background

Infection with high-risk human papillomavirus type 16 (HPV-16) is associated with cervical oncogenesis. The prevalence of HPV-16 lineages has not been well investigated in Africa. There is a lack of comprehensive epidemiological data on the distribution and oncogenic potential of the different HPV-16 lineages and sub-lineages in these populations. Through systematic review and meta-analysis, we aimed to determine the prevalence of HPV-16 lineages among women in Africa between 1997 and 2023.

Method

Eight databases were searched, and studies that identified HPV-16 positive women with distinct HPV-16 lineages, and those that clearly defined an African population, were included. The eight databases include PubMed, PubMed Central, BVS, Embase, Cochrane Library, Scopus, Web of Science, and EBSCOhost. This protocol is registered in PROSPERO (CRD42023437748) and followed the PRISMA guidelines.

Results

Out of the 2698 studies that were screened, 12 studies met the inclusion criteria and were included in the systematic review and meta-analysis. Analysis of these 12 studies identified yielded 1209 HPV-16-positive samples based on distinct diagnostic tests. These samples were classified as lineage A (n = 556), lineage C (n = 278), lineage B (n = 235), and lineage D (n = 140). Overall, lineage A variants were found in 11 studies (556 samples), establishing an overall prevalence of 58% (95% CI: 0.31–0.80; I² = 96.0%; τ² = 2.924; p < 0.0001). Lineage C variants were identified in 8 studies (278 samples), with an overall prevalence of 39% (95% CI: 0.21–0.60; I² = 84.1%; τ² = 1.362; p < 0.0001). Lineage B variants were identified across 7 studies (235 samples), with a prevalence of 20% (95% CI: 0.09–0.37; I² = 90.4%; τ² = 1.109; p < 0.0001). Lineage D variants were identified across 5 studies (140 samples), and its overall prevalence was 10% (95% CI: 0.05–0.19; I² = 89.1%; τ² = 0.672; p < 0.0001).

Conclusion

Women in Africa exhibit a varied distribution of the four HPV-16 lineages. The high prevalence of HPV-16 lineage A in the entire population calls for developing targeted interventions to maximize impact and effectiveness. Future studies that are standardized and include defined geographic region, cytological details, and individual patient data should be conducted to enhance robust findings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13643-025-02891-3.

Introduction

Persistent infection by oncogenic human papillomavirus (HPV) is considered the etiologic cause of up to 90–95% of cervical cancers [1, 2]. HPVs have been classified into low-risk and high-risk types depending on their carcinogenic potential [3, 4]. Among the oncogenic HPV types, HPV-16 is generally responsible for about half of all cervical cancer cases, followed by HPV-18, and, together, they account for up to 70–76% cervical cancers [5–8].

There is significant genetic variability among the HPV-16 strain with multiple variants identified based on differences in their nucleotide sequences [9–12]. Nucleotide variations in HPV-16 are known to change amino acids and also impact the expression of proteins ultimately influencing the cervical cancer development [9]. Specific genetic variations within HPV-16 types have been implicated in both cervical carcinogenesis and viral transmission and persistence. These variants may influence the virus’s ability to cause cancer and its spread making them important factors to consider in HPV research and prevention strategies [13, 14].

Initially, HPV-16 genetic variants were classified into five branches according to the geographical and phylogenetic relatedness based on the long control region (LCR) of its genomic sequence: European, Asian-American (AA), Asian (As), African-1 (Af-1), and African 2 (Af-2) [15, 16]. The nomenclature of HPV-16 genetic variants was revised in 2012 [9]. The updated classification system categorizes variants into lineages (1–2% sequence difference) and sub-lineages (0.5–1% sequence difference) based on whole genome sequencing. The range boundaries of sequence divergence are empirically defined based on genetic variability observed across the complete genomes of HPV-16 variants, hence fixed [17]. This more precise classification is expected to provide better understanding of HPV-16 diversity and its potential implication for viral transmission, persistence, and carcinogenesis [9]. The variant lineages are thus described by letters (A, B, C, D) while the sub-lineages by numbers [1–4]. Supplementary Table 1 shows the comparison of the former and current nomenclature [18].

Research shows that approximately 23.7% of women in the general population harbor cervical HPV infection at any specific point in time [19]. The prevalence of any HPV infection varies greatly according to geographical distribution and has been shown to have a strong correlation with cervical cancer occurrences [20, 21]. For instance, HPV infections influence the progression of abnormal cervical cytology [13]. In many instances, cervical abnormalities are often considered indicators of cervical cancer development. Several studies have shown that HPV-16 lineages A (A1, A2) and D have been detected in many regions of North Africa. However, lineages B and C have been reported more frequently in sub-Saharan Africa [9, 22]. Few systematic reviews and meta-analyses have been conducted in the past that focused on diverse HPV-16 lineages by gender, geographical location, cytological condition, and linkage to genital cancers [18, 23–25]. Therefore, this is the first systematic review and meta-analyses to exclusively focus on HPV-16 lineages among women in Africa. Knowledge of the HPV-16 lineages and sub-lineages among women in Africa contributes significantly to achieving the Sustainable Development Goal (SDG) 3 that focuses on health and well-being. Primarily, the findings will promote targeted public health intervention such as defined treatment approaches and vaccination strategies towards elimination of cervical cancer in Africa. The knowledge will also assist in optimization of screening programs by prioritizing resources and screening efforts towards the most at-risk population. Therefore, this study identifies the prevalence and distribution of HPV-16 lineages among African women through a systematic review and meta-analysis of the literature.

Methodology

To identify the prevalence of HPV-16 lineages among women in Africa, this systematic review with meta-analysis followed the proposed protocol registered in the International Prospective Register of Systematic Reviews (PROSPERO, CRD42023437748). The systematic review was also guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA 2020) (Supplementary Table 3) [26].

Search strategy

The search strategy was designed by a librarian (A. P.) upon discussion with I. A., D. O., M. T., and F. O. to identify studies reporting HPV-16 lineages among women in Africa. A comprehensive literature search was conducted on September 26, 2023, and October 8, 2023, across eight databases: PubMed, PubMed Central, BVS/BIREME/MEDLINE and LILACS, Embase, Cochrane, Scopus, Web of Science, and EBSCOhost (MEDLINE Complete, MEDLINE, Academic Search Premier, CINAHL, Food Science Sources, Open Dissertations, Dentistry & Oral Sciences Sources, SPORTDiscuss, Business Source Premier). A combination of defined MeSH terms in the title or abstract or key words without restricting the language of articles were used:

• (Women OR Girls OR Girl OR Woman OR "Women's Groups" OR "Women Groups" OR "Women's Group"): (Female OR Females); ("Human papillomavirus 16" OR "HPV16" OR "HPV16s" OR "human papillomavirus type 16 HPV16" OR "Human papillomavirus type 16" OR "Human papillomavirus—16" OR "human papillomavirus type 16 HPV 16"OR"HPV-16" OR "HPV 16"):

• "Genetic Variation" OR "Genetic Variations" OR "Variations, Genetic" OR "Variation, Genetic" OR "Diversity, Genetic" OR "Diversities, Genetic" OR "Genetic Diversities" OR "Genetic Diversity":

• Prevalence OR Prevalences and (Different names used to refer to Africa including Each African country name). A comprehensive search strategy was employed, incorporating the names of all African countries as keywords. This approach maximized the identification of relevant studies, even those that may not explicitly reference Africa in their work. The exact search strategy is comprehensive and has therefore been attached in Supplementary Table 2.

Study selection

Three reviewers were involved in the study selection process. First, all duplicate articles were removed. Subsequently, the first reviewer (I. A.) and the second reviewer (D. O.) independently screened the titles, abstracts, and key words of the articles for eligible studies using the Rayyan QCRI program [27]. Rayyan is a validated tool that facilitates the screening of articles for systematic review [28]. The screening was based on inclusion and exclusion criteria. The inclusion criteria were studies that identified HPV-16 genetic variants among female subjects only, and the variants had to be clearly defined. This meant that the studies to be included must have detailed nucleotide sequence data or specific mutational profile that gives room for accurate classification of HPV-16 into their respective lineages and sub-lineages. The second criteria were that studies must have been carried out in Africa. It meant that the studies must have been conducted within the geographical boundaries of the African continent. Exclusion criteria included case reports, conferences, reports, and in vitro/in vivo studies. This meant that non-peer-reviewed documents such as conference abstracts, government or organizational reports, and other gray report that do not undergo the rigorous peer review process would not be included. This would ensure validity and reliability of the findings. Another exclusion criterion entailed studies done outside of Africa and studies including males only. Further, studies where lineages could not be determined with the available data were also eliminated. Any conflict was resolved with a participation of a third reviewer (F. O). Upon selection of the articles in the first phase, the first and second reviewer (I. A. & D. O.) read the articles in their entirety and further selected them according to the inclusion and exclusion criteria. Any further conflict was resolved by a third reviewer (F. O.). All the three reviewers during the screening phases were blinded.

Data extraction

Data extraction included study characteristics such as author name, year of publication, and location of the study. Clinical characteristics extracted included specimen collection method, biological specimen type and condition, molecular technique, genomic region used for variant assessment, number of HPV-16 specimens, and HPV-16 lineages identified. Methodological characteristics included study design, setting, sample size, and year of sample collection. All these characteristics were first entered as raw data in Excel sheets. The data to be used for meta-analyses was further extracted in duplicate by I. A. and D. O. to ensure quality. Due to inconsistencies in HPV-16 variant nomenclature found in several manuscripts, a standardized nomenclature system was devised using whole genome sequence data. The papillomavirus Episteme database was used for reference [29]. Therefore, all HPV-16 lineages were classified using the current nomenclature of HPV-16 lineage (A, B, C, D).

Assessment of risk of bias

The methodological quality of the studies included in the systematic review with meta-analysis was measured using risk-of-bias tools (robvis) for observational cohort and cross-sectional studies (https://www.riskofbias.info/welcome/robvis-visualization-tool). The tool assessed the quality of each study based on seven domains that focused on different biases such as bias due to confounding, bias arising from measurement of exposure, bias in selection of participants, bias due to post-exposure intervention, bias due to missing data, bias arising from measurement outcome, and bias in selection of reported results. Each individual study underwent individual-level judgements for each domain, and the judgement was reported using traffic light plots classified as high (red), low (green), and some concerns (yellow) where “high” means significant methodological flaws and biases in the individual study, “low” means the study is sound and free from significant biases that could affect the validity of the findings, while “some concerns” means there is insufficient information to make a valid definitive judgement.

Statistical analysis

Statistical analyses were conducted in R statistical software version 4.4.0. The Microsoft Excel sheet containing data was imported into R for analysis. An inverse variance-weighted random effects meta-analysis of logit-transformed proportions was performed to estimate pooled prevalence of HPV lineage with the corresponding 95% confidence intervals (CI). The use of logit transformation was necessitated by the need to improve normality of the sample proportions. The random-effects model of analysis was assumed since the studies identified were observational in nature and they had both clinical and methodological variabilities. Prevalence was calculated as a ratio of the number of HPV-16 lineage (each lineage A, B, C, D) identified to the total number of HPV-16 lineage samples drawn from the population. Prevalence of HPV-16 lineage was displayed in forest plots. Misclassification was accounted for by elimination of all HPV-16 lineage samples that could not be fully identified either as A, B, C, or D.

The evidence for the presence of magnitude of heterogeneity between studies was determined and assessed using the Cochran’s Q test and quantified using the I² test statistic. If the degree of heterogeneity was above 75%, the studies were considered heterogeneous [30]. The p-value is associated with observed heterogeneity. We then performed a subgroup analysis to assess if heterogeneity differed by study setting, molecular technique used, and biological sample used. We used funnel plots for visual inspection of publication bias. Symmetrical plots were interpreted as no publication bias, whereas publication bias was indicated when plots were asymmetrical. Further, we assessed for small study effect using contour-enhanced funnel plots. In the presence of publication bias, we performed trim and fill analysis. Sensitivity analysis was conducted excluding studies that did not report events on specific lineage types for each HPV lineage.

Results

The process of systematically reviewing articles to identify the prevalence of HPV-16 lineages among women in Africa is summarized in Fig. 1. The initial search strategy produced 2698 articles. Of those, 1829 were duplicates and therefore excluded, leaving a total of 869 articles eligible for screening. Of these, 22 articles were eligible for full review, and 12 studies that met the inclusion criteria were selected for inclusion in the systematic review and meta-analysis.

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of the literature search

Risk of bias

After assessing each individual study, an overall assessment on the quality of the study was automatically generated by the tool. In this review, an overall low risk of bias was assessed from the 12 studies as shown in the results (see supplementary Table 5).

Summary of characteristics of the 12 eligible studies

Epidemiological understanding of each study is important in order to understand the quantitative information on HPV-16 lineages. The nonquantitative data (qualitative data) was also extracted and has been used in the synthesis of the quantitative data findings. The data has been classified as clinical and methodological characteristics of the study.

Table 1 summarizes the clinical and methodological characteristics of the samples that were included in the study. The author and year of publication have been included for clarity. All samples in the study were drawn from the cervix of women. The characteristic noted included year of sample collection. Out of 12 studies, four studies did not specify the year of sample collection. Only eight studies had clear indication of the years ranging between 1999 and 2023. The geographical region where the study samples were drawn has been included. In these instances, the geographical regions were divided into two: North Africa and sub-Saharan Africa. North Africa included countries that fall in the northernmost part of the Africa continent primarily bordering Mediterranean Sea and the Red Sea, while sub-Saharan Africa includes countries that are located south of Sahara Desert [31]. Some studies were crosscutting in both regions. The study setting was also reclassified to a single country when samples were collected from one country whereas multinational when samples were drawn from several countries. The molecular techniques that have been used to detect HPV-16 lineages have also been outlined and reclassified to those that used PCR and Sanger sequencing or other methods. The biological samples have also been described according to the time collection in reference to the study. There are studies that used samples stored in the biorepositories, while other studies used freshly taken samples. The genomic region of HPV-16 used in the sequencing also varies from one study to another. Genomic region refers to specific segment of DNA within the HPV-16. In this case, the genomic regions are key in identification of the HPV-16 lineage. Some studies used E6 region only, while others combined E6, E7, L1, LCR, and other genomic regions. The HPV-16 lineages established (A, B, C, D) have also been indicated for each study. A total of 1677 samples were obtained, of which 468 included samples that had ambiguous classifications and were thus excluded from the analysis. Ambiguous classification meant that the identified HPV-16 lineage could not be independently classified as lineage A, B, C, and D. Other characteristics of these studies have been included in Supplementary Table 6. They have not been included in the main study characteristics because of their inconsistencies.

Table 1.

Showing clinical and methodological characteristics of the study

Author	Year	Year of sample collection	Molecular technique	Genomic region	Study setting	Region	Biological sample source	Sample	A	B	C	D	Misclassified
Yamada T	1997	NA	Taq PCR/Sanger	E6/LCR	Multinational	North/sub-Saharan Africa	Biorepository	64	6	39	19	0	0
J. J., T	2006	1999	PCR/Sanger	E6	Single country	Sub-Saharan	Freshly taken	93	52	14	24	3	0
Lei Y. J	2011	2007	PCR/Sanger	LCR	Single country	Sub-Saharan	Freshly taken	11	10	0	0	0	1
Krenn Hrubec	2011	2007	Dot blot hybridization assays/Sanger	E1, E2, E6, E7, L2	Single country	North Africa	Biorepository	64	41	0	0	0	23
Qmichou	2013	2012	Nested PCY/Sanger	E6	Single country	North Africa	Biorepository	103	60	7	24	12	0
Cornet I	2013	N/A	PCR/Sanger	E6	Multinational	North/sub-Saharan	Biorepository	688	198	0	0	64	426
Moussavou Boundzanga	2016	2014	Nested PCR/Sanger	E6/LCR	Single country	Sub-Saharan	Freshly taken	14	2	0	0	0	12
Ghedira	2016	2004	RT-PCR/Sanger	Ll1	Single country	North Africa	Freshly taken	56	48	0	6	2	0
Awua	2017	N/A	Nested PCR/Sanger	LCR	Single country	Sub-Saharan	Biorepository	18	0	1	17	0	0
Clifford, G. M	2019	N/A	NGS	E6/L1/LCR	Multinational	North/sub-Saharan	Biorepository	521	133	167	162	59	0
Iglesias	2022	2017	Clinical Array Technology (CLART) HPV4 assay/Sanger	L1/LCR	Single country	Sub-Saharan	Freshly taken	7	2	1	4	0	0
Kaba G	2023	2020	Nested PCR/Sanger	E6/E7	Single country	Sub-Saharan	Freshly taken	36	4	6	22	0	4

Overall prevalence of HPV-16 lineages

The overall prevalence of HPV-16 lineages among women in Africa was determined. There were high levels of heterogeneity between studies for all HPV-lineage variants. Overall, lineage A variants were found in 11 studies (556 samples), establishing an overall prevalence of 58% (95% CI: 0.31–0.80; I² = 96%; τ² = 2.924; p < 0.0001). Lineage C variants were identified in 8 studies (278 samples), with an overall prevalence of 39% (95% CI: 0.21–0.60; I² = 84.1%; τ² = 1.362; p < 0.0001). Lineage B variants were identified across 7 studies (235 samples), with a prevalence of 20% (95% CI: 0.09–0.37; I² = 90.4%; τ² = 1.109; p < 0.0001). Lineage D variants were identified across 5 studies (140 samples), and its overall prevalence was 10% (95% CI: 0.05–0.19; I² = 89.1%; τ² = 0.672; p < 0.0001). The forest plots for random-effects meta-analysis of the prevalence of HPV-16 lineages have been presented below. The boxes indicate the effect size of the studies (prevalence), and the whiskers indicate its CI for the corresponding effect size. In the case of diamonds, their size indicates the size of the effect, and their length indicates CIs (Figs. 2, 3, 4, and 5).

Fig. 2.

Forest plots for random-effects meta-analysis of the prevalence of HPV-16 lineage A

Fig. 3.

Forest plots for random-effects meta-analysis of the prevalence of HPV-16 lineage B

Fig. 4.

Forest plots for random-effects meta-analysis of the prevalence of HPV-16 lineage C

Fig. 5.

Forest plots for random-effects meta-analysis of the prevalence of HPV-16 lineage D

Subgroup analysis of HPV-16 lineages

Subgroup analysis was also performed to determine the weighted prevalence in comparison to three parameters, namely the study setting, molecular technique, and the biological sample used. Regarding study setting, lineages A and C were majorly reported in a single country setting with pooled prevalence of 69% (95% CI: 0.37–0.89) and 44% (95% CI: 0.18–0.74), respectively. Lineages B and D were reported in a multinational setting with pooled prevalence of 46% (95% CI: 0.21–0.19) and 17% (95% CI: 0.08–0.34). There are eight studies whose study setting was a single country. Only three studies belonged to a multinational setting.

Most samples used in this study were drawn from the biorepository. Specifically, lineages B, C, and D reported high pooled prevalence in samples drawn from biorepository: 21% (95% CI: 0.06–0.55), 44% (95% CI: 0.15–0.78), and 15% (95% CI 0.09–0.25), respectively. For instance, in lineage B, four studies reported the use of samples from the biorepository, while only three had samples that were freshly taken. Only the majority of lineage A samples were freshly taken: 60% (95% CI: 0.26–0.87). Six studies reported using samples from one country only, while five studies had samples that had been freshly taken.

In terms of molecular techniques used for HPV-16 lineage detection, Sanger sequencing was the most common method of HPV-16 lineage detection. Most samples from lineages A, B, C, and D used PCR/Sanger sequencing for analysis 57% (95% CI: 0.29–0.82), 18% (95% CI: 0.06–0.41), 39% (95% CI: 0.16–0.69), and 9% (96% CI: 0.03–0.22), respectively. For example, eight studies that identified lineage A used PCR and Sanger sequencing. There was only one study that reported use of next-generation sequencing. Heterogeneity did not seem to decline with sub-group analysis (Table 2).

Table 2.

Subgroup analysis based on study setting, molecular technique, and biological samples in the included studies

Variable	No. of studies	Sample size		Pooled prevalence (95% CI)	I2	τ2	P
Lineage A
Freshly taken	5	200	118	0.60	87.3%	2.5386	< 0.0001
Biorepository	6	1009	438	0.57	98%	4.5180	< 0.0001
Total	9	1209	556	0.58	96.0%	2.9241	< 0.0001
Single country	8	362	219	0.69	85.3%	2.8917	< 0.0001
Multinational	3	847	337	0.33	98.9%	2.8739	< 0.0001
Total	11	1209	556	0.58	96.0%	2.9241	< 0.0001
PCR/Sanger	8	640	380	0.57	92.9%	2.5784	< 0.0001
NGS	1	521	133	-	-	-	-
Other assay/Sanger	2	48	43	0.84	90.4%	12.8698	0.0012
Total	11	1209	556	0.58	96.0%	2.9241	< 0.0001
Lineage B
Freshly taken samples	3	200	21	0.16	-	0-	0.8796
Biorepository	4	1009	214	0.21	93.8%	2.1597	< 0.0001
Total		1209	235	0.20	90.4%	1.1092	< 0.0001
Single country		362	29	0.12	25.7%	0.1312	< 0.0001
Multinational		847	206	0.46	94.8%	0.6780	< 0.0001
Total		1209	235	0.20	90.4%	1.1092	< 0.0001
PCR/Sanger	5	640	67	0.18	93.5%	1.5238	< 0.0001
NGS	1	521	167	-	-	-	-
Other assay/Sanger	1	48	1	-	-	-	-
Total	7	1209	235	0.20	90.4%	1.1092	< 0.0001
Lineage C
Freshly taken	5	200	56	0.36	89.9%	1.5556	< 0.0001
Biorepository	4	1009	222	0.44	80.3%	2.1407	0.0016
Total	9	1209	278	0.39	84.5%	1.3622	< 0.0001
Single country	6	362	97	0.44	88.9%	2.2819	< 0.0001
Multinational	2	847	181	0.31	-	0.8183	0.8183
Total	8	1209	278	0.39	84.5%	1.3622	< 0.0001
PCR/Sanger	6	640	112	0.39	88.3%	2.109	< 0.0001
NGS	1	521	162	-	-	-	-
Other assay/Sanger	1	48	4	-	-	-	-
Total			Lineage D
Freshly taken	2	200	5	0.03	-	0.9097	< 0.0001
Biorepository	3	1009	135	0.15	91.4%	0.2612	< 0.0001
Total	5	1209	140	0.10	89.1%	0.6724	< 0.0001
Single country	3	362	17	0.06	66.2%	0.4745	0.0520
Multinational	2	847	123	0.17	95.4%	0.4113	< 0.0001
Total	5	1209	140	0.10	89.1%	0.6724	< 0.0001
PCR/Sanger	4	640	81	0.09	88.5%	1.0205	< 0.0001
NGS	1	521	59	-	-	-	-
Other assay/Sanger	1	48	0	-	-	-	-
Total	6	1209	140	0.10	89.1%	0.6724	< 0.0001

Sensitivity analysis

By excluding studies that did not report detection of specific lineage types for each HPV-lineage subtype, the results showed a decline in heterogeneity in the analysis of all the four HPV-16 lineages. The prevalence of HPV-16 lineages A and C declined, while for B and D it remained the same. Prevalence of lineage A is 46% (95% CI: 0.23–0.71; I² = 93.4%; τ² = 2.783; p < 0.0001), B is 20% (95% CI: 0.09–0.37; I² = 90.6%; τ² = 1.120; p < 0.0001), C is 37% (95% CI: 0.21–0.57; I² = 83.0; τ² = 1.171; p < 0.0001), and D is 10% (95% CI: 0.08–0.12; I² = 51.3%; τ² = 0.0096; p < 0.0838) (Supplementary Table 4).

Discussion

Comprehensive data on the prevalence of HPV-16 lineages is crucial for estimating the burden of cervical cancer and possibly maximizing the impact of HPV vaccination strategies and treatment approaches in cervical cancer prevention. To the best of our knowledge, this is the first comprehensive systematic review and meta-analysis to explore the distribution of HPV-16 lineages among women in the African context. The results reveal an uneven distribution of different HPV-16 lineages among women in Africa. HPV-16 lineage A is the most commonly detected variant, followed by variants C and B. In this meta-analysis, variant D was the least detected among women in Africa [8, 9, 32–34].

The findings of the study underscore the global dominance of lineage A. The comprehensive prevalence of HPV-16 lineage A is likely attributed to its evolutionary history and transmission dynamics. It is hypothesized that this lineage emerged during Neanderthal interbreeding and the subsequent out-of-Africa migrations, leading to its increased prevalence in both African and non-African populations [35]. Previous studies have supported the claim. Ferreira et al. conducted a systematic review with a meta-analysis to establish the genetic variants of HPV-16, as well as the geographical and anatomical distribution in men globally [18]. The study revealed that lineage A was the most prevalent variant in terms of anatomical site and geographical distribution [18]. From their analysis, all samples of lineage A variants were most commonly detected independent of the country where samples were obtained. The highest prevalence of lineage A variants was however observed in Europe (86.59%), Asia (90.28%), and North America (84.59). However, there were no samples on HPV-16 lineages from the African population in this study [18]. Moreover, in a study that was done in Northeastern Brazil to establish the role of HPV-16 among cervical carcinoma samples, the results indicated a high prevalence of HPV-16 lineage A among confirmed cervical carcinoma cases at 49% and variants B and C at 4% [36]. This corroborates the findings of the present systematic review of the high prevalence of HPV-16 lineage A globally.

Lineages B and C have in the past been reported to be prevalent in Africa [37–39]. This study revealed a significant presence of the two lineages in Africa. However, because of misclassifications that led to the elimination of some samples, the accuracy of the overall reported prevalence of lineages B and C in this work could be compromised. The misclassification can be attributed to the fact that there are five studies that failed to accurately distinguish whether the HPV-16 genetic variant was B or C and only classified them as non-European lineages or non-A lineages [9, 40–44]. The failure to accurately distinguish the lineages is primarily due to limitations in genomic regions analyzed and insufficient resolutions of the sequencing methods. The studies mentioned focused on specific regions of the HPV genome such as E6 and LCR regions which in many instances may not provide sufficient phylogenetic resolution that can distinguish closely related lineages such as B and C. The sequencing techniques used in these studies may also not have been comprehensive enough to detect detailed genetic differences as evident in Table 1. The lack of distinction meant they were not included in this systematic review with meta-analysis, hence the possible low prevalence of HPV-16 lineages B and C.

The low prevalence of HPV-16 lineage D in the African population corroborates past findings on the geographical distribution of HPV-16 lineages. Studies show that lineage D is mainly prevalent in Latin America with its sub-lineages D1, D2, and D3 significantly being associated with cervical cancer in northeastern Argentina and Guatemala [45, 46].

Another major finding from the study was the common use of PCR/Sanger sequencing as a method of molecular analysis. In all the studies, a high prevalence of lineages A, B, C, D was reported in samples that utilized PCR/Sanger sequencing. Currently, HPV-16 lineage determination is mainly possible using sequencing. Sanger sequencing is often considered a gold standard method in HPV-16 lineages identification [47]. Nevertheless, it is expensive and labor-intensive and may not be easily accessible in resource-limited settings such as Africa. This could suggest the reason for the limited number of studies on HPV-16 lineage detections that have been conducted in Africa as evident in the systematic review.

Clinical implication

Africa presents a unique landscape of HPV-16 lineages with distinct distribution that could potentially affect the efficacy of HPV vaccines. The varying prevalence of HPV-16 lineages A, B, C, and D within different population may contribute to development of vaccine resistance. About half of African countries have integrated HPV vaccination into their national immunization programs [48]. However, given the diverse epidemiological landscape of HPV-16 lineage in Africa, a tailored approach to HPV vaccination is crucial. Further, the bivalent vaccine which targets HPV-16 and HPV-18 is already in use in some countries in Africa. Specifically, the vaccine targets the L1 protein of HPV-16 which is a major capsid protein that forms virus-like particles and induces type-specific neutralizing antibodies. While the vaccine is considered highly effective against HPV-16, it does not differentiate between the various lineages and sub-lineages of HPV-16. Developing targeted HPV vaccine that address specific HPV-16 lineages could enhance vaccine efficiency and public health impact. However, this would call for more comprehensive understanding of geographical distribution of HPV-16 lineages as well as ongoing surveillance to monitor possible changes in the prevalence of lineages in different populations.

Strength and limitation

This is the first to the best of our knowledge to explore the distribution of HPV-16 lineages among women in Africa. This is crucial because of the high burden of cervical cancer in Africa despite global efforts by WHO towards cervical cancer elimination [49]. The study is based on a thorough systematic literature search that focuses on each African country independently to find out eligible studies that have explored HPV-16 lineages which are crucial causative factors of cervical cancer.

However, this meta-analysis revealed substantial heterogeneity among the studies, as evidenced by the high I² statistic. This finding is not uncommon in prevalence studies [50, 51]. To address this heterogeneity, the researchers attempted to conduct subgroup analyses based on biological samples, study setting, and molecular techniques. However, these efforts were unsuccessful in reducing the heterogeneity, suggesting the presence of heterogeneity within the subgroups themselves.

One potential source of heterogeneity could be the HIV status of the women, which was only reported in two studies. Subgrouping by HIV status could have potentially reduced heterogeneity, but the limited data prevented this analysis. Additionally, the source and condition of the biological samples could contribute to the variation. Biorepositories are known to house samples collected from diverse settings with limited associated patient data, which can introduce significant heterogeneity. Additionally, samples stored in biorepositories are susceptible to degradation over time, compromising their integrity when compared to freshly taken samples. To mitigate these challenges, rigorous quality control measures, standardized sample collection protocols, and comprehensive data management systems are essential to produce homogeneous studies.

Again, it was not possible to do subgroup analysis by cytological conditions of women and distinct region where the samples were sourced. This was majorly attributed to the lack of detailed individual patient data in the included studies. As noted in a previous writing, subgroup analyses of subsets of participants within studies are uncommon in systematic reviews due to the limited availability of detailed patient data in published studies [52]. Further, while the researchers attempted to categorize studies by geography, inconsistencies in the geographical scope of the studies limited the feasibility of meaningful subgroup analysis by region. For instance, some studies were conducted exclusively in sub-Saharan Africa and others spanning both North and sub-Saharan Africa; thus, it was not feasible to perform a meaningful subgroup analysis by geographical region. Other inconsistencies were noted in the year of sample collection, genomic area of analysis, and study design, hindering further analysis.

However, the reduction of heterogeneity observed in the sensitivity analysis suggests that the inclusion of studies which did not have specific lineages could have possibly introduced variability in the pooled prevalence. The findings highlight the importance of consistent and reliable methods of lineage analysis. To address this issue, future studies should collect more detailed patient data, including HIV status and cytological condition, to enable more precise subgroup analyses. Additionally, standardized data collection and reporting practices across studies would facilitate more robust meta-analysis.

The study findings are subject to potential selection bias. Specifically, studies reporting high prevalence rates of certain HPV-16 lineages or statistically significant variations in prevalence across regions or populations may be more readily published than those with low prevalence or null results. This tendency towards publishing positive findings could lead to an overestimation of the actual prevalence of specific HPV-16 lineages across Africa. Furthermore, the inclusion of studies primarily published in English may have resulted in the exclusion of relevant research documented in other languages. Consequently, the meta-analysis might present a biased prevalence estimate that does not fully represent the diversity of findings across the African continent.

Conclusion

Cervical cancer remains a significant global health threat, particularly in Africa; hence, understanding the epidemiological landscape of HPV-16 lineages is crucial for its elimination. This study provides valuable insights into the prevalence of HPV-16 lineages among African women. By examining different populations of women from various settings, the study sheds light on the distribution and burden of specific HPV-16 lineages across Africa. The high prevalence of HPV-16 lineage A in the entire population calls for developing targeted interventions focusing on HPV vaccination and defined treatment approaches to maximize their impact and effectiveness. However, there is also a need to carry out well-designed large-scale studies in several regions of Africa to draw refined comparisons.

Authors’ contributions

IA, conceptualization, methodology, formal analysis, investigation, data curation, writing — original draft, and review and editing. MTF, conceptualization, methodology, investigation, data curation, writing — original draft, review and editing, supervision, and project administration. DO, FO, and SSA, methodology, investigation, data curation, and review and editing. RVML, formal analysis and review and editing. JB, review and editing, supervision, project administration, and funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

VLIR-UOS program, Jaramogi Oginga Odinga University of Science and Technology. The funding body had no role in the design of the study, data collection, analysis and interpretation of the data, or writing of the manuscript.

Data availability

All data during study are included in this manuscript and supplementary files.

Declarations

Competing interests

The authors declare that they have no competing interests.