Cervical Cancer Burden in Africa: An Integrative Review of Epidemiology, Plant Medicine, and AI-Driven Innovation

Abstract

Cervical cancer imposes a heavy burden in low- and middle-income countries and persistently high incidence and mortality in Sub-Saharan Africa due to unequal access to prevention and care. Although human papillomavirus (HPV) is a major driving factor, the risk is amplified by HIV co-infection, socioeconomic barriers, limited infrastructure, low health literacy, and behavioral factors, leading to frequent late-stage diagnosis, low screening coverage, and poor survival. Recent expansion of HPV vaccination programs since 2019 has achieved high coverage in some countries, but continent-wide rates remain below targets amid hesitancy, inequities, and systemic weaknesses. Thus, there is a critical need for an integrated synthesis of multidisciplinary evidence that reflects the peculiarities and diversities of African health systems. Moreover, high mortality rates underscore the need for integrated, context-specific evidence to strengthen vaccination, screening, and management strategies in line with WHO global elimination targets. Plant medicine promises to complement mainstream health delivery in low resource settings. Emerging artificial intelligence (AI) tools show strong potential to improve screening accuracy, reduce subjectivity and resource needs, and enhance early detection in resource-limited settings. This review synthesizes multidisciplinary data on economic impact, epidemiology, risk factors, prevention, diagnostic challenges, plant medicine and AI innovations to guide equitable progress in diverse African contexts.

1. Introduction

Cervical cancer remains a major global health challenge, with a disproportionate burden in low- and middle-income countries. In 2020, approximately 604,000 new cases and 342,000 deaths were recorded worldwide, making it a leading cause of cancer morbidity and mortality among women [1]. Although organized screening and widespread HPV vaccination have substantially reduced disease burden in high-income countries, comparable progress remains limited across much of Africa [2,3]. Sub-Saharan Africa continues to experience some of the world’s highest incidence and mortality rates, reflecting persistent inequalities in preventive and curative healthcare access [4].

Persistent infection with high-risk HPV, particularly HPV-16 and HPV-18, is the principal driver of cervical carcinogenesis [5]. In Africa, this biological risk is compounded by HIV infection, which promotes HPV persistence and malignant progression [3,6]. Socioeconomic disadvantage, limited health literacy, inadequate healthcare infrastructure, and restricted access to vaccination and screening further increase vulnerability. Behavioral and environmental factors, including early sexual debut, multiple sexual partners, tobacco use, prolonged contraceptive use, and co-infections, may further influence disease progression and outcomes.

Late-stage diagnosis remains common because of inadequate screening coverage and weak referral systems. Population-based registry data indicate variable but generally unfavorable incidence patterns, with some regions experiencing stable or increasing rates [7]. Mortality remains high, with late presentation contributing substantially to poor survival [8]. Across surveys conducted between 2000 and 2020, screening coverage remained low in most African countries, with few eligible women ever screened and fewer receiving repeat testing [9]. Consequently, delays in diagnosis and limited access to specialized treatment continue to undermine survival despite the availability of effective interventions [4,8].

Prevention efforts have increasingly focused on expanding HPV vaccination and screening. Since 2019, more African countries have introduced national HPV vaccination programmes through international partnerships and improved vaccine supply [10]. However, coverage remains below global targets, with persistent disparities affecting rural communities, out-of-school adolescents, and other underserved populations. Vaccine hesitancy, misinformation, financial constraints, and weak health systems continue to challenge programme sustainability [10].

Emerging technologies may provide additional opportunities to strengthen cervical cancer control. Artificial intelligence (AI) has demonstrated potential for improving screening and diagnostic accuracy while reducing processing time, dependence on specialized personnel, and subjective interpretation [11]. Plant-derived medicines also represent a potentially valuable complementary resource, particularly in low-resource settings, although clinical evidence remains limited.

Despite extensive research, evidence concerning cervical cancer in Africa remains fragmented across epidemiology, prevention, treatment, economics, and emerging therapeutic approaches. This review therefore integrates multidisciplinary evidence on the epidemiological and economic burden of cervical cancer, current treatment capacity, and the evolving roles of medicinal plants and AI in research and clinical practice. Integrating prevention, early detection, effective treatment, and responsible technological innovation could accelerate progress toward reducing cervical cancer incidence and mortality across Africa.

2. Methodology

This review used a structured narrative literature search and evidence synthesis approach to integrate evidence on the epidemiology, molecular pathogenesis, risk factors, prevention, diagnosis, treatment, economic burden, medicinal plants, and artificial intelligence (AI) applications relevant to cervical cancer in Africa. The methodology was designed to capture both Africa-specific evidence and high-quality global evidence needed to interpret regional findings and contemporary clinical and technological developments.

2.1. Literature Databases and Information Sources

Literature searches were conducted using PubMed/MEDLINE, Scopus, and Web of Science Core Collection. Google Scholar served as a supplementary source for identifying relevant publications, reports, and articles not readily retrieved through the principal bibliographic databases. Additional authoritative sources were consulted where appropriate for epidemiological estimates, clinical guidance, health system information, and economic indicators, including the World Health Organization (WHO), the International Agency for Research on Cancer (IARC), GLOBOCAN, the Global Cancer Observatory, the World Bank, FIGO, ASCO, and NCCN. Reference lists of relevant reviews and primary studies were also screened manually to identify additional pertinent literature.

2.2. Search Period and Timeframe

The principal literature search covered publications from 2000 through 2026, with older landmark studies retained when they provided foundational evidence for HPV-mediated carcinogenesis, cervical cancer screening, treatment, or other concepts directly relevant to the review. Priority was given to recent evidence, particularly publications from 2015 onward, for epidemiological estimates, prevention, treatment, medicinal-plant research, and AI applications.

2.3. Search Strategy and Keywords

Search terms were combined using the Boolean operators AND and OR and were adapted to the syntax of each database. The principal concepts and keywords spanned several domains. For disease and epidemiology, terms included cervical cancer, cervical carcinoma, cervical neoplasia, cervical intraepithelial neoplasia, and CIN. The geographical focus incorporated Africa, sub-Saharan Africa, and individual countries including Ghana, Nigeria, Kenya, Uganda, Tanzania, Senegal, Malawi, Cameroon, South Africa, Mozambique, Morocco, Egypt, Tunisia, Eswatini, and the Democratic Republic of the Congo. For HPV and pathogenesis, searches combined human papillomavirus, HPV, HPV16, HPV18, E6, E7, p53, pRB, viral integration, and carcinogenesis. Prevention and screening terms included vaccination, HPV vaccine, screening, self-sampling, VIA, cytology, HPV testing, and prevention. Diagnosis and treatment searches drew on diagnosis, cytology, histopathology, biopsy, MRI, PET, radiotherapy, brachytherapy, chemotherapy, chemoradiotherapy, surgery, FIGO, ASCO, and NCCN. Economic-burden terms comprised economic burden, health expenditure, treatment cost, direct cost, indirect cost, productivity, and financial burden. Medicinal-plant searches used medicinal plants, phytotherapy, phytochemicals, plant-derived, herbal, extract, and named genera including Artemisia, artesunate, Curcuma, Moringa, Annona, Vernonia, and Bidens. For artificial intelligence, terms included artificial intelligence, AI, machine learning, deep learning, radiomics, computer vision, predictive modelling, and clinical decision support.

Searches were conducted both as broad topic searches and as focused searches linking cervical cancer with Africa and each individual evidence domain. For medicinal plants, plant names were additionally cross-searched against cervical cancer, HeLa, SiHa, cervical intraepithelial neoplasia, apoptosis, cytotoxicity, xenograft, animal study, and clinical trial. For AI, searches combined cervical cancer with cytology, histopathology, whole-slide imaging, MRI, PET, radiomics, HPV status, molecular data, screening, diagnosis, treatment planning, prognosis, and clinical decision support.

2.4. Eligibility and Inclusion Criteria

Eligible sources comprised peer-reviewed original research articles, systematic reviews, meta-analyses, major narrative reviews, and authoritative guidelines or reports directly relevant to cervical cancer. Priority was given to studies addressing cervical cancer or cervical precancer, with particular emphasis on African populations, African health systems, or evidence with direct relevance to African settings, and to studies reporting epidemiology, incidence, mortality, HPV prevalence or genotype distribution, risk factors, screening, vaccination, diagnosis, treatment, health system capacity, costs, or economic consequences. For medicinal plants, eligible studies investigated plant extracts or defined plant-derived compounds with evidence relevant to cervical cancer or cervical precancer, including mechanistic in vitro studies, animal studies, and human clinical studies. For AI, eligible studies evaluated machine learning, deep learning, computer vision, radiomics, predictive modelling, or clinical decision-support applications in cervical cancer screening, diagnosis, treatment planning, response assessment, prognosis, or related molecular or clinical prediction tasks. Publications had to be available in English and sufficiently detailed to permit assessment of study findings and relevance. Landmark older studies were eligible when they established fundamental mechanisms or clinical principles that remain relevant to contemporary cervical cancer research.

Studies were excluded where they represented duplicate publications or multiple records reporting the same study, in which case the most complete or authoritative report was retained. Studies unrelated to cervical cancer or cervical precancer, or studies in which cervical cancer findings could not be separated from other malignancies, were excluded, as was non-peer-reviewed material lacking clear scientific or institutional authority, including commercial promotional material, opinion pieces, news articles, and publications lacking sufficient methodological or outcome information. Studies for which the full article or sufficiently reliable bibliographic information could not be established were excluded from evidence synthesis. For country-level economic comparisons, figures lacking a clear definition, reference year, or internationally comparable methodology were excluded from quantitative comparison.

2.5. Study Selection and Evidence Synthesis

Retrieved records were screened by title and abstract for relevance, followed by examination of the full text of potentially eligible publications. Reference lists of key papers were reviewed for additional studies. Evidence was organized thematically according to the major domains of the manuscript, namely molecular pathogenesis, epidemiology and risk factors, prevention and screening, diagnosis and treatment, economic burden, medicinal plants, and AI. Greater weight was given to systematic reviews, meta-analyses, authoritative international guidelines, large observational studies, randomized clinical trials, and well-characterized experimental studies. For medicinal plants, findings were categorized according to evidence level so that in vitro observations were not presented as evidence of clinical efficacy. For economic indicators, internationally standardized definitions and common reference years were preferred, including WHO current health expenditure as a percentage of GDP.

2.6. Quality and Relevance Considerations

Because this manuscript is a multidisciplinary narrative review rather than a formal systematic review or meta-analysis, no pooled quantitative effect estimate was generated and no formal risk-of-bias instrument was applied across all evidence domains. Nevertheless, study design, population, geographical relevance, sample size where applicable, methodological transparency, consistency of findings, and level of evidence were considered when interpreting the literature. Particular caution was applied to extrapolation from laboratory or animal studies to human treatment, and to extrapolation from non-African populations to African settings.

2.7. Data Extraction and Presentation

Relevant information was extracted qualitatively from eligible sources and synthesized into the corresponding sections. Extracted information included study setting, population or experimental model, disease stage where applicable, intervention or technology, principal outcome, evidence level, and major limitations. An economic-burden table uses a consistent WHO-defined health-expenditure indicator to improve comparability between countries.

3. Molecular Pathogenesis of Cervical Cancer

Cervical cancer is primarily driven by persistent infection with high-risk human papillomavirus (HPV), particularly HPV-16 and HPV-18, which account for most cases globally and in Africa [12,13]. Integration of oncogenic HPV DNA into the host genome can sustain viral oncogene expression and promote malignant transformation [14]. Central to this process is deregulated expression of the viral E6 and E7 oncogenes, which disrupt key tumor-suppressor pathways governing cell-cycle control and genomic stability [15]. Persistent infection facilitates clonal expansion and progression from cervical intraepithelial neoplasia to invasive carcinoma, explaining the near-universal association between cervical cancer and high-risk HPV [16]. Nevertheless, HPV infection alone is insufficient for malignancy, indicating an essential contribution from host molecular alterations (Figure 1).

Figure 1. Simplified mechanistic model of HPV-driven cervical carcinogenesis. Persistent infection with high-risk HPV, particularly HPV16 and HPV18, facilitates viral genome integration into the host genome and sustained expression of the viral oncogenes E6 and E7. E6 recruits E6-associated protein (E6-AP), promoting ubiquitination and proteasomal degradation of the tumour suppressor p53, thereby impairing DNA-damage-induced cell-cycle arrest and apoptosis. E7 binds and functionally inactivates retinoblastoma protein (pRb), releasing E2F transcription factors and promoting unscheduled S-phase entry and cellular proliferation. Disruption of these tumour-suppressor pathways is accompanied by p16^INK4a^ accumulation, increased Ki-67 expression, genomic alterations, altered cellular metabolism, and progressive cervical dysplasia, ultimately contributing to invasive cervical cancer. This mechanistic sequence reflects the molecular processes described in the manuscript.

HPV-mediated cervical oncogenesis involves a host of viral oncoproteins such as E6 and E7 (Figure 1). E6 promotes E6-associated protein (E6-AP)-mediated ubiquitination and proteasomal degradation of p53, impairing DNA-damage-induced cell-cycle arrest and apoptosis [17]. Concurrently, E7 binds and inactivates retinoblastoma protein (pRb), releasing E2F transcription factors and inducing unscheduled S-phase entry and sustained cellular proliferation [18,19]. Together, E6-mediated p53 disruption and E7-mediated pRb inactivation represent the central molecular axis of HPV-associated cervical carcinogenesis [13,15].

pRb disruption also induces accumulation of p16^INK4a^, making it a widely used biomarker of transcriptionally active high-risk HPV infection [20]. Increased Ki-67 expression reflects heightened proliferation and impaired cell-cycle regulation, while combined p16^INK4a^/Ki-67 assessment can improve differentiation of low-grade lesions, high-grade neoplasia, and invasive carcinoma [21,22].

Host genomic alterations further contribute to malignant progression. Recurrent mutations in PIK3CA, EP300, FBXW7, and PTEN, together with altered immune-regulatory pathways such as increased PD-L1 expression, cooperate with HPV-mediated oncogenic activity to promote tumor development and immune evasion [23,24]. Epigenetic dysregulation, including aberrant DNA methylation of tumor-suppressor promoters and methylation changes associated with HPV integration, further alters gene expression and may provide opportunities for molecular detection and disease stratification [25].

In Africa, considerable diversity in circulating HPV lineages may influence viral persistence and carcinogenic risk [26,27]. HIV-associated immunosuppression further promotes persistent HPV infection and accelerates cervical disease progression, contributing to the increased incidence and severity observed in affected populations [3,6]. These molecular interactions highlight the need for region-specific research to better characterize cervical carcinogenesis and develop more effective diagnostic, preventive, and therapeutic strategies in African populations [6,27].

4. Economic Impact

Cervical cancer is a major global health and economic concern, with an estimated 662,301 cases and 348,874 deaths in 2022, ranking fourth among cancers affecting women worldwide [28,29,30]. Africa bears a disproportionate share of the global cervical cancer burden relative to its share of the world’s female population, reflecting persistent deficiencies in prevention, diagnosis, and treatment [28,31]. The resulting economic burden includes healthcare expenditure, household costs, productivity losses, and premature mortality.

Direct costs include diagnosis, surgery, chemotherapy, and radiotherapy, while transportation and accommodation contribute additional non-medical expenses. Indirect costs arise from lost income, reduced productivity, and premature death [29]. Late diagnosis increases treatment complexity and expenditure, particularly in low- and middle-income settings, while inadequate insurance and high out-of-pocket payments amplify household financial hardship [29].

The burden varies across regions. In West Africa, Ghana records more than 3000 new cases annually, while nationwide HPV vaccination would require approximately US$11–15 million annually, equivalent to 11–15% of its national immunization budget [32]. In 2022, only 31% of girls in the WHO African Region received their first dose of HPV vaccine by age 15, despite 26 countries having introduced HPV vaccination into their national immunization programmes [33]. In Nigeria, high out-of-pocket payments for cancer care contribute to catastrophic household expenditure [34].

Eastern Africa experiences some of the continent’s highest incidence rates. Eastern Africa experiences some of the continent’s highest incidence rates [35]. Kenya experiences financial pressures associated with late-stage disease and costly radiotherapy and chemotherapy [36]. Tanzania incurs significant indirect costs from lost income and travel for specialized care [37].

Southern Africa bears particularly high per-capita losses. Eswatini’s annual societal burden is estimated at US$19 million [38], while South Africa faces escalating oncology costs associated with HIV-related disease and late presentation [39]. Malawi and Mozambique similarly face substantial treatment and productivity losses linked to limited screening and delayed diagnosis [35,37,40].

In Central Africa, weak cancer surveillance likely underestimates the burden in Cameroon and the Democratic Republic of Congo [35,37]. Morocco, Egypt, and Tunisia generally report lower incidence but continue to experience treatment and productivity costs, particularly among underserved populations [30,41]. The EMENA cost-effectiveness framework illustrates how vaccination costs relative to GDP per capita can guide prevention investment [42,43].

Overall, cervical cancer imposes substantial economic costs across Africa (Table 1). Expanding vaccination, screening, and timely treatment is therefore both a public health priority and a strategic economic investment.

Table 1. Economic burden context of cervical cancer in selected African countries.

Country	Current Health Expenditure (% of GDP), 2021	Key Economic Burden Drivers	Reference
Ghana	3.86%	High HPV vaccination financing needs; public-sector treatment subsidies	[32,37]
Nigeria	4.08%	High out-of-pocket spending and catastrophic expenditure	[34,37]
Senegal	4.44%	Limited screening coverage; late-stage presentation	[34,35,37]
Uganda	5.47%	Limited oncology capacity and late presentation	[34,35,37]
Kenya	4.57%	Radiotherapy and chemotherapy costs; late-stage disease	[36,37]
Tanzania	3.47%	Travel and referral costs to tertiary centres	[37]
Eswatini	7.41%	High per-capita cervical cancer burden and treatment costs	[37,38]
South Africa	8.66%	Large oncology expenditure; HIV-associated disease and late presentation	[37,39]
Malawi	7.52%	Late-stage treatment and productivity losses	[34,35,37]
Cameroon	4.34%	Underfunded screening; high mortality-related losses	[35,37]
Democratic Republic of the Congo	3.87%	Weak health financing; informal care costs	[35,37]
Mozambique	8.65%	Vaccine delivery costs; limited surveillance data	[35,37,40]
Morocco	5.45%	Moderate treatment costs; disparities in early detection	[37,42,43]
Egypt	4.61%	Sustained public oncology spending	[37,41,42]
Tunisia	7.34%	Health system capacity and treatment costs	[37,42,43]

Note: Values are current health expenditure (CHE) as a percentage of GDP for 2021, taken from the WHO Global Health Expenditure Database (December 2025 release), indicator GHED_CHEGDP_SHA2011 [37].

5. Epidemiology and Risk Factors

Cervical cancer remains a major cause of cancer morbidity and mortality among African women, with sub-Saharan Africa experiencing a disproportionately high burden despite its largely preventable nature [44,45]. Persistent gaps in primary prevention, screening, and timely treatment continue to sustain this burden.

Unlike high-income countries, where incidence has generally declined, cervical cancer incidence across several sub-Saharan African countries has remained stable or increased, reflecting inadequate population coverage and weaknesses in screening implementation [7]. Mortality remains high, largely because of late-stage diagnosis and limited treatment infrastructure [46].

High-risk HPV infection, particularly HPV-16 and HPV-18, remains the principal cause of cervical cancer, with considerable regional variation in HPV prevalence and genotype distribution [47]. HIV further amplifies this risk by promoting persistent HPV infection and accelerating progression from precancerous lesions to invasive disease. Consequently, cervical cancer incidence and mortality are particularly elevated in high-HIV-prevalence settings, including Eastern and Southern Africa [46,47].

Screening uptake remains a major weakness. Low coverage has been documented across sub-Saharan Africa, with substantial disparities according to wealth, education, and urban-rural residence [48]. Even among Kenyan women frequently accessing maternal and child health services, screening uptake remains inadequate [49]. Among women living with HIV, screening uptake is also disproportionately low despite their elevated risk [50,51]. Major barriers include limited services availability, cost, stigma, fear, low awareness, and inadequate integration of screening into HIV care [44,52,53].

Socioeconomic and behavioral factors further shape disease risk. Poverty, rural residence, limited education, cultural misconceptions, and fear of social rejection reduce preventive service utilization [44,48,54]. In Ghana, limited awareness among young women and adolescents, together with inadequate access to health information, remains an important concern [55,56]. Early sexual debut, multiple sexual partners, and high parity increase HPV exposure and cervical cancer risk, while limited condom use and inadequate male partner involvement may sustain transmission [45,57,58].

Precancerous lesions are common, particularly among women living with HIV, representing important opportunities for early intervention [59]. However, delayed diagnosis and inadequate treatment access continue to increase mortality [46]. Detailed incidence, mortality, case numbers, and prevalence are presented in Figure 2, Figure 3 and Figure 4. Country-level estimates are from the IARC Global Cancer Observatory (GLOBOCAN 2022) [28,31]. The dataset provides cancer burden estimates by country, including case numbers, mortality, crude rates and age-standardised rates.

Figure 2. Global mortality and incidence rates of cervical cancer. Rates are expressed per 100,000 women. The map depicts the worldwide spatial distribution of cervical cancer incidence and mortality, highlighting significant geographic disparities in disease burden. Incidence rates, illustrated through choropleth shading, are most pronounced in sub-Saharan Africa, Latin America, Eastern Europe, and certain areas of South and South-East Asia, whereas mortality rates, indicated by proportional red circles, are notably high in sub-Saharan Africa and other low- and middle-income regions. Conversely, North America, Western Europe, and Australia demonstrate significantly lower incidence and mortality rates, indicative of the efficacy of structured screening initiatives, widespread HPV vaccination, and enhanced access to prompt treatment. The observed disparity between incidence and mortality in various regions underscores enduring inequalities in early detection and healthcare access, emphasising global evidence that cervical cancer is predominantly preventable yet continues to inflict disproportionate mortality in areas with the weakest health systems [1,28,31,60].

Figure 3. Global incidence rates and number of cases of cervical cancer. Rates are expressed per 100,000 women. The map depicts the worldwide distribution of cervical cancer incidence rates and total case numbers, emphasizing significant regional disparities in disease burden. Sub-Saharan African nations, along with regions in South and South-East Asia, Latin America, and certain areas of Eastern Europe, demonstrate comparatively high incidence rates and significant case numbers, indicative of the interplay between heightened risk and population density. Conversely, North America, Western Europe, and Australia exhibit lower incidence rates, despite significant case numbers in certain nations, largely due to efficient population-based screening, widespread HPV vaccination, and timely access to treatment. The simultaneous occurrence of elevated incidence and substantial case numbers in numerous low- and middle-income nations highlights ongoing inequalities in prevention and healthcare systems, emphasizing the global public health necessity to enhance equitable cervical cancer screening, vaccination, and treatment initiatives [1,28,31,60].

Figure 4. Global prevalence and incidence rate of cervical cancer. Rates are expressed per 100,000 women. The map illustrates a global representation of the prevalence and incidence rates of cervical cancer. Countries are depicted in varying shades of purple to represent the global prevalence rate per 100,000 women, with darker hues indicating higher prevalence (e.g., sub-Saharan Africa) and lighter shades denoting lower prevalence (e.g., the majority of Europe and North America). Superimposed on this choropleth are proportional red circles denoting the incidence rate of cervical cancer—larger circles indicate higher rates. The aggregation of substantial circles in areas like East and Southern Africa, South Asia, and certain regions of South America underscores the elevated incidence of cervical cancer, corroborating findings that the burden of cervical cancer is predominantly found in low- and middle-income nations with restricted access to HPV vaccination and screening services. This spatial distribution highlights enduring global health disparities: high-income countries exhibit lower incidence due to more extensive screening and vaccination, whilst low-income regions bear a greater illness burden due to inadequate health systems [1,28,31,60].

6. Cervical Cancer Treatment and Access in Africa

Cervical cancer treatment aims to prevent progression, reduce mortality, and preserve quality of life, with earlier intervention associated with better survival [61,62]. Treatment selection depends on disease stage, patient health, and potential effects on fertility, hormonal balance, and immune function [63,64,65,66,67]. Surgery, radiotherapy, chemotherapy, immunotherapy, and targeted therapy constitute the principal modalities and are guided by FIGO, ASCO, and NCCN recommendations (Figure 5) [68,69]. Because treatment effectiveness and toxicity vary, individualized planning, surveillance, fertility preservation, and psychosocial support remain essential [69,70,71,72]. Advances in precision medicine and targeted treatment may further improve efficacy while reducing toxicity [73].

Despite these advances, access across Africa remains constrained by inadequate vaccination and screening, limited surgical and radiotherapy capacity, and fragmented health systems [34,35,74]. WHO’s African strategy prioritizes HPV vaccination among girls aged 9–14 years and screening among women aged 30–49 years [42]. By February 2024, only 28 of 47 countries had incorporated HPV vaccination into national programmes, with regional first-dose coverage at approximately one third of the target; Ethiopia, Seychelles, and Uganda exceeded 90% coverage [42]. Rwanda’s decentralized vaccination model and Zambia’s introduction of vaccination in 2019, achieving approximately 70% two-dose coverage by 2020, demonstrate feasible approaches [42].

Screening remains uneven, with many countries relying on VIA rather than HPV DNA testing. Screen-and-treat programmes have been implemented in Benin, Cameroon, Côte d’Ivoire, and Senegal, while Malawi integrates VIA and cryotherapy into reproductive health services. Nevertheless, screening coverage among women aged 30–49 years remains approximately 11% regionally, and rural-urban disparities persist despite integration into services in Kenya, Tanzania, and Mozambique [35,42]. Kenya’s national cervical cancer programme has expanded screening and introduced HPV vaccination, but implementation remained constrained by low screening coverage, limited HPV-testing availability, and weak linkage to treatment [75]. Community-based interventions have also demonstrated the potential to expand access to cervical cancer prevention services in Nigeria; a prospective study in rural communities in southeast Nigeria found that house-to-house education by trained community health educators substantially increased cervical cancer screening uptake [76]. Although 34 Member States provide some screening and 16 have adopted HPV DNA testing, continued reliance on VIA and cytology limits diagnostic sensitivity [42]. Africa CDC’s planned Centers of Excellence in Rwanda, Morocco, and Zambia seek to strengthen vaccination and screening toward elimination, although late-stage presentation remains common [42,74].

Surgical management is similarly limited. Although early-stage disease can be treated curatively with radical hysterectomy and lymphadenectomy, delayed presentation and shortages of trained gynecologic oncologists restrict access [42,77]. Radical hysterectomy is available in tertiary centers in Uganda and Zambia but not routinely in Eswatini and Rwanda, illustrating persistent specialist shortages [42,77]. Access to oncologic surgeons, operating theatre constraints, inadequate imaging, and infrastructure deficits remain important barriers. Mentorship programmes in the Democratic Republic of the Congo have strengthened surgical capacity among general gynecologists, while Malawi has used neoadjuvant chemotherapy followed by surgery where radiotherapy is unavailable [77,78]. Limited postgraduate training and very low surgeon-to-patient ratios further constrain capacity [79].

Chemotherapy access is also inconsistent. Platinum-based chemoradiotherapy is widely recommended; yet drug stockouts; treatment costs; and access barriers frequently interrupt care [80,81]. National protocols exist in countries such as The Gambia and Kenya; but detailed tertiary guidance remains limited elsewhere [82]. East African data indicate that fewer than half of patients with common cancers receive guideline-concordant chemotherapy [34]. Although Uganda’s Cancer Institute provides chemotherapy; only about two thirds of WHO Africa Region Member States report chemotherapy services [42]. Procurement weaknesses; out-of-pocket costs; limited essential drug availability; shortages of oncology professionals, and inadequate supportive care further restrict treatment [42,74,80].

Radiotherapy and brachytherapy remain the most constrained modalities. Fewer than 40% of WHO Africa Region Member States provide radiotherapy, and available data indicate that Egypt and South Africa hold nearly half of the continent’s brachytherapy capacity [42,83]. In Ethiopia, prolonged delays in accessing radiotherapy remain a major challenge; a longitudinal study reported a median interval of 477 days between diagnosis and receipt of radical-dose radiotherapy among patients with cervical cancer [84]. Equipment failures, unreliable power, and shortages of physicists and technologists further reduce operational capacity, while Central Africa faces particularly severe brachytherapy deficits [80,83]. Planned Centers of Excellence offer an important opportunity to strengthen regional capacity, but sustained financing, infrastructure development, and workforce investment will be essential to translate treatment guidelines into equitable care [74,83].

Figure 5. Stage-adapted management pathway for cervical cancer (FIGO IA2-IIA) with risk-stratified postoperative treatment. This schematic shows the guideline-concordant treatment algorithm for FIGO stage IA2-IIA cervical cancer. Surgical candidates undergo sentinel lymph node biopsy followed by radical hysterectomy or trachelectomy with lymphadenectomy. Postoperative care depends on risk: low-risk disease is observed; intermediate-risk features (large tumor size, deep stromal invasion, lymphovascular invasion) prompt adjuvant radiotherapy; high-risk features (positive nodes, parametrial invasion, positive margins) warrant chemoradiotherapy. Non-surgical candidates receive definitive radiotherapy or concurrent chemoradiotherapy. This risk-adapted approach balances oncologic control against overtreatment, reflecting real-world decisions in resource-variable settings, including African health systems, where FIGO, ASCO, and NCCN-aligned protocols must fit infrastructure capacity [69,85].

Figure 5. Stage-adapted management pathway for cervical cancer (FIGO IA2-IIA) with risk-stratified postoperative treatment. This schematic shows the guideline-concordant treatment algorithm for FIGO stage IA2-IIA cervical cancer. Surgical candidates undergo sentinel lymph node biopsy followed by radical hysterectomy or trachelectomy with lymphadenectomy. Postoperative care depends on risk: low-risk disease is observed; intermediate-risk features (large tumor size, deep stromal invasion, lymphovascular invasion) prompt adjuvant radiotherapy; high-risk features (positive nodes, parametrial invasion, positive margins) warrant chemoradiotherapy. Non-surgical candidates receive definitive radiotherapy or concurrent chemoradiotherapy. This risk-adapted approach balances oncologic control against overtreatment, reflecting real-world decisions in resource-variable settings, including African health systems, where FIGO, ASCO, and NCCN-aligned protocols must fit infrastructure capacity [69,85].

7. Medicinal Plants and Cervical Cancer

Cervical cancer remains a major cause of cancer morbidity and mortality among women in sub-Saharan Africa, largely driven by persistent high-risk HPV infection and inadequate access to screening and treatment. Medicinal plants and plant-derived compounds have attracted increasing attention as potential sources of adjunctive or locally delivered interventions. The existing literature spans mechanistic studies using cervical cancer cell models, animal experiments, and a smaller but increasingly important body of human research. The distinction between these evidence levels is critical because promising cytotoxicity in cell culture does not, by itself, establish therapeutic efficacy in humans. HeLa cells remain an important model for preliminary investigation of plant-derived anticancer activity [86,87,88,89].

Among the plant-derived candidates considered in the present review, Artemisia annua and its derivatives, particularly artesunate and dihydroartemisinin, have the clearest translational pathway. Artemisinin derivatives have been associated with ROS-mediated apoptosis, cell-cycle arrest, ferroptosis, anti-angiogenic activity, and inhibition of invasion and metastasis [90]. Iron-dependent activation of the endoperoxide moiety can disturb redox homeostasis, while modulation of MAPK, NF-κB, and PI3K/Akt/mTOR signalling has been implicated in anti-angiogenic and antiproliferative effects [91,92]. Importantly, evidence extends beyond cell culture. In a cervical cancer xenograft model, artesunate enhanced the response of HeLa-derived tumours to irradiation, with greater tumour suppression, apoptosis, and cell-cycle effects than irradiation alone, although the response was less evident in SiHa-derived tumours [93]. This model-dependent effect provides experimental support for investigating artesunate as a potential radiosensitizing adjunct rather than as an established stand-alone treatment.

The translational relevance of Artemisia annua is further supported by human studies of intravaginal artesunate for cervical precancer. A first-in-human phase I study of artesunate vaginal inserts in women with biopsy-confirmed CIN2/3 reported that the treatment was generally safe and well tolerated; histologic regression to CIN1 or less was observed in 19 of 28 evaluable participants, while HPV clearance occurred in a subset of women whose lesions regressed [94]. A subsequent Kenyan programme has extended investigation of self-administered intravaginal artesunate to women with cervical precancer, including women living with HIV, with the objective of evaluating HPV-related outcomes, safety, adherence, acceptability, and lesion response [95]. These studies involve cervical precancer rather than established invasive cervical cancer, and therefore should be interpreted as evidence of translational potential rather than proof of anticancer efficacy.

Other plants show encouraging mechanistic activity, but their evidence remains predominantly preclinical. Vernonia amygdalina demonstrates apoptosis-related activity involving p53, JNK, caspase-9, PTEN, and XBP1, while isolated phytosterols have been associated with S-phase arrest and PI3K/AKT/mTOR modulation [96,97]. Antiproliferative activity has also been reported in hepatocellular carcinoma cells [98]. Xylopia aethiopica has been associated with mitochondrial apoptosis, caspase activation, loss of mitochondrial membrane potential, altered Bax/Bcl-2 balance, and reduced migration in cervical cancer models [99,100,101]. Bidens pilosa and related species contain flavonoids, terpenoids, alkaloids, and polyacetylenes with reported effects on apoptosis, cell-cycle regulation, oxidative stress, and kinase signalling [102,103,104]. Cytotoxic or antiproliferative activity against cancer cell lines has also been reported for extracts of Sarcocephalus pobeguinii, Dialium guineense, Kalanchoe pinnata, Manilkara zapota, Ceropegia rupicola, Acokanthera schimperi, Elephantorrhiza elephantina and Jatropha variegata [105,106,107,108,109,110]. Phytochemical and pharmacological profiling of Piptadeniastrum africanum, together with anti-inflammatory constituents from Rhaphiostylis beninensis and Dorstenia psilurus, further illustrates the chemical diversity of candidate plants [111,112,113]. These findings justify further investigation but do not yet provide sufficient in vivo or clinical evidence for inclusion among the leading candidates.

Similarly, Moringa oleifera has shown activity through crude extracts and nano-formulations, including modulation of NF-κB and PI3K/AKT signalling and enhancement of oncolytic vesicular stomatitis virus activity through Bax and STAT1-associated mechanisms [114,115,116]. Annona muricata has been linked to apoptosis, cell-cycle arrest, mitochondrial dysfunction, ROS generation, and acetogenin-mediated inhibition of mitochondrial complex I [101,117,118]. However, much of the cervical cancer evidence for both plants are based on cellular or computational studies. Rosmarinus officinalis and Otostegia integrifolia likewise show antiproliferative, oxidative-stress, inflammatory, and apoptosis-related effects in experimental models [119,120,121,122,123,124,125]. These observations remain useful for identifying candidate compounds and mechanisms, but they should not be equated with demonstrated clinical benefit.

A further consideration is that plant-derived interventions have already entered human studies, but clinical results have not been uniformly positive. A randomized phase II trial of Polyphenon E, a decaffeinated green-tea catechin preparation derived from Camellia sinensis, in women with persistent high-risk HPV infection and CIN1 found the preparation to be acceptable and well tolerated but did not demonstrate an improvement in HPV or CIN1 clearance compared with placebo [126]. This finding illustrates why clinical validation, rather than mechanistic plausibility alone, must determine which plant-derived candidates progress toward clinical use. In contrast, a clinical intervention using a standardized extract of Curcuma longa in women with persistent low-grade cervical squamous intraepithelial neoplasia reported arrest or regression of lesions over 12 weeks, with no major clinical or biochemical safety signal in the small study population [127]. The study provides human evidence of biological activity, but its small sample size and study design warrant cautious interpretation.

The principal barriers to clinical translation include variability in plant species and chemotypes, differences in extraction procedures, inconsistent concentrations of active constituents, uncertain pharmacokinetics, limited pharmacovigilance, inadequate product standardization, and uncertainty regarding dosing and long-term safety [128]. Few studies on Bidens species include clinical participants [129], and controlled human evidence is likewise insufficient for Moringa [130] and Annona muricata [131]. Although Moringa has immunomodulatory and pro-apoptotic properties, controlled human evidence remains insufficient [130]. Concerns surrounding prolonged use of Annona muricata also reinforce the need for systematic safety evaluation, particularly in view of reported neurotoxicity concerns and the absence of credible human efficacy evidence for cervical cancer [131]. Accordingly, future work should prioritize chemically characterized extracts, defined active constituents, reproducible formulations, pharmacokinetic and toxicological assessment, appropriate animal models, and adequately powered clinical trials [128]. Plant cell and tissue culture offers one route to reproducible, scalable production of defined metabolites independent of field variability, as shown by the enhanced yields of 2,3-dihydrobenzofuran and 3-epilupeol from Ageratina pichinchensis cell suspensions scaled up from shake flasks to an airlift bioreactor [132]. On the basis of the evidence currently available, Artemisia annua derivatives, particularly artesunate, and selected preparations of Curcuma longa have progressed furthest toward human evaluation, while most of the other plants reviewed remain at the preclinical stage.

8. Application of Artificial Intelligence in Detection and Treatment of Cervical Cancer

Artificial intelligence (AI) is increasingly influencing oncology by extracting clinically relevant patterns from complex biomedical data to support clinical decision-making (Figure 6) [133,134]. In cervical cancer, machine learning and deep learning are being explored for screening, diagnosis, treatment planning, and prognostic assessment, offering greater consistency and scalability than human-dependent approaches [133,135]. By reducing inter-observer variability and processing large volumes of cytological, histopathological, radiological, and molecular data, AI may be particularly valuable where specialist capacity is limited [11]. Importantly, AI is intended to augment rather than replace clinical expertise (Figure 6) [134].

Figure 6. Artificial intelligence–driven framework for cervical cancer detection, treatment planning, and prognostication. This schematic illustrates the integration of multimodal data, including cytology and histopathology images, MRI/PET imaging, molecular and genomic profiles, and clinical variables, within an AI analytical framework. Machine learning, deep learning, radiomics, predictive modeling, and clinical decision-support tools synthesize these datasets to support: (i) screening and early detection through automated cytology classification and dysplasia identification; (ii) diagnostic enhancement through lesion localization and histopathological grading; (iii) radiotherapy planning through tumor segmentation, organ-at-risk delineation, and toxicity prediction; and (iv) precision treatment stratification through prediction of responses to chemoradiation and immunotherapy. Collectively, these applications may improve early detection, treatment personalization, toxicity reduction, follow-up, and survival prediction. Clinical implementation requires data quality, transparency, external validation, ethical oversight, and equitable deployment.

Image-based applications represent one of the most developed areas. Deep learning models can analyze cervical cytology and histopathology images to identify morphological abnormalities associated with dysplasia and malignancy, including nuclear atypia and altered nuclear-to-cytoplasmic ratios [135,136]. Automated analysis can improve consistency, reduce fatigue-related errors, and support earlier lesion detection [11]. Digital pathology further enables whole-slide assessment, facilitating lesion localization and grading while potentially extending specialist diagnostic capacity in resource-constrained settings [134,137].

Radiomics provides another emerging application. By extracting high-dimensional imaging features related to tumor heterogeneity, vascularity, and texture, AI can identify patterns that may inform diagnosis, staging, treatment response, and prognosis [138,139]. In cervical cancer, such models may reveal imaging biomarkers of tumor aggressiveness and improve tumor delineation, with potential implications for radiotherapy planning and monitoring [134,139].

AI also supports radiotherapy through automated segmentation of tumors and organs at risk, potentially reducing planning time and inter-observer variability. Machine learning can additionally predict treatment-related toxicity using patient-specific characteristics, supporting more individualized and risk-aware treatment planning [134,140]. Beyond radiotherapy, integration of molecular, clinical, and imaging data may facilitate prediction of treatment response and patient stratification, including responses to chemoradiation and emerging immunotherapies, although cervical cancer-specific validation remains limited [133,141].

Prognostic modeling represents another promising frontier. AI can integrate clinical, pathological, and imaging variables to predict survival, recurrence, and treatment response, potentially identifying nonlinear relationships that conventional models overlook [141,142]. Such predictions may support individualized follow-up, treatment decisions, counseling, and resource allocation [134,142].

Despite these opportunities, implementation faces substantial challenges, including data quality, algorithmic bias, transparency, limited digitization, inadequate computational infrastructure, and insufficient prospective validation [133,134]. Representative datasets, regulatory oversight, workforce training, and integration into clinical workflows are essential for equitable adoption, particularly in African health systems [34]. With appropriate validation and governance, AI could become a powerful adjunct to cervical cancer prevention, diagnosis, treatment, and long-term outcome prediction [142].

9. Conclusions and Future Prospects

Cervical cancer remains a major cause of cancer-related mortality among women in Africa. Its persistent burden reflects the interaction of high-risk HPV infection, HIV co-infection, inadequate vaccination and screening, socio-cultural barriers, treatment limitations, and fragile health systems. Addressing these challenges requires integrated, context-specific approaches that move beyond isolated interventions toward coordinated prevention, early detection, and equitable treatment.

Future progress will depend on expanding HPV vaccination and accessible screening, including self-sampling approaches, while strengthening public education and health system capacity. Advances in understanding the molecular mechanisms of cervical carcinogenesis may further enable more precise and effective therapies. Medicinal plants also represent a potentially valuable source of affordable therapeutic agents, but their clinical relevance will depend on rigorous pharmacological characterization, standardization, mechanistic investigation, and appropriately designed clinical trials.

Artificial intelligence presents an emerging opportunity to transform cervical cancer control, particularly through automated screening, digital cytology, image-based diagnostics, and predictive risk assessment. However, meaningful implementation in African settings will require investment in digital infrastructure, workforce development, data quality, ethical governance, and safeguards against algorithmic bias.

Cervical cancer elimination in Africa will require sustained political commitment, stronger research partnerships, and equitable investment across the prevention-to-treatment continuum. Integrating established public health measures with advances in medicinal plant research and responsible digital innovation could accelerate progress toward an Africa in which cervical cancer is no longer a leading cause of preventable death.