Luận văn cervical cancer screening by co testing method for vietnamese women from 25 to 55 years old a cost effectiveness analysis

Literature review

Epidemiology of cervical cancer

Human papillomavirus (HPV) is a significant cause of various tumors and cancers, particularly cervical cancer (CC) in women While HPV is associated with oropharyngeal, anal, and vulvar cancers, these types were not included in this study due to their multiple etiologies The proportion of these cancers attributed to HPV varies, and although there are over 200 recognized HPV genotypes, only a select few are linked to cancer The International Agency for Research on Cancer (IARC) has identified 12 high-risk HPV types, including types 16 and 18, as carcinogenic to humans.

HPV is primarily transmitted through sexual contact, but most infections regress within one to two years due to the body's immune response, often clearing the virus within five years Typically, HPV infections are asymptomatic, with clearance more common in young adults However, persistent infections with high-risk HPV types and elevated viral loads can lead to the development of preneoplastic lesions and cervical cancer The preclinical phase of cervical cancer can span 15 to 20 years in women with normal immune systems, while it may only take 5 to 10 years in those with weakened immune systems This extended timeline underscores the importance of cervical cancer screening for early detection, which has proven effective over the years Many women with HPV and precancerous conditions can recover and clear the infection following treatment.

Burden of cervical cancer in the world and Southeast Asia region

In 2017, the Global Burden of Disease Study reported that approximately 601,000 women were diagnosed with cervical cancer globally, leading to 260,000 deaths Although detected cases slightly decreased to 570,000 in 2018, fatalities rose to 311,000, making cervical cancer the fourth most common cancer among women Worldwide, the likelihood of a woman developing cervical cancer in her lifetime is estimated at 1 in 65, with significantly higher risks in countries with low social-demographic indices (SDI) In these low SDI countries, the risk increases to 1 in 40 women, compared to 1 in 106 women in high SDI nations.

Cervical cancer was a major cause of cancer-related deaths among women globally in 2007, and a decade later, it remains prevalent in 50 countries, ranking as the leading cause of cancer deaths for women in 39 of those nations Furthermore, from 2007 to 2017, both the incidence and mortality rates of cervical cancer rose by 19%.

In 2017, cervical cancer accounted for 7.3% of global cancer cases, ranking third in cancer-related deaths and contributing to a 15% increase in Disability-Adjusted Life Years (DALYs) lost According to the International Agency for Research on Cancer, projections for 2020 indicated that cervical cancer incidence would position it as the fourth most common cancer worldwide.

Figure 1.1: Estimated number of new cases in 2020, females, all ages - Source: GLOBOCAN 2020 - Graph production:

Global Cancer Observation Retrieved from: https://gco.iarc.fr/today

Recent data highlights significant disparities in the incidence and mortality rates of cervical cancer between high-income and lower middle-income countries, primarily attributed to the differences in HPV vaccination and screening program coverage.

Figure 1.2: Estimated number of new cases in 2020, High income (left) and Low middle income (right), females, all ages

- Source: GLOBOCAN 2020 - Graph production: Global Cancer Observation Retrieved from: https://gco.iarc.fr/today

In 2008, cervical cancer was the leading cause of cancer-related deaths among women in Southeast Asia, and a decade later, it remained a significant public health issue, ranking second in morbidity and mortality across all age groups in the region In contrast, western Asia reported the lowest incidence rates, with fewer than 5 cases per 100,000 women The high prevalence of HIV among individuals aged 15 to 49 and the elevated risk of HIV transmission in Southeast Asia may account for the stark differences in cervical cancer rates between these subregions.

Burden of cervical cancer in Viet Nam

In Vietnam, cervical cancer was the fifth most prevalent cancer and the seventh leading cause of cancer-related deaths among both men and women in 2017, according to global statistics Recent estimates from the International Agency for Research on Cancer indicate a concerning prevalence of cervical cancer in Vietnam from 2020 onward.

Figure 1.3: Estimate number of prevalent cases - Source: GLOBOCAN 2020 - Graph production: Global Cancer

Observation Retrieved from: https://gco.iarc.fr/today

A literature review by Lam Duc Tam reveals significant variations in HPV infection rates across provinces, with Thua Thien Hue at 0.9% and Ho Chi Minh City at 19.97%, a difference of nearly 20 percentage points Additionally, research by Arbyn et al indicates that cervical cancer ranks among the top three cancers affecting women under 45 in 176 countries, including Vietnam This prevalence is linked to the high percentage of women aged 30 and older.

A study conducted by Tran Kim Ngoc and colleagues revealed a significant increase in cervical cancer incidence among women in Vietnam, particularly peaking in the 55 to 59 age group, after rising sharply from ages 30 to 34 This trend highlights the urgent need for HPV awareness and prevention strategies in the region, especially given that 49 cases of HPV infection were documented in several provinces.

Between 1999 and 2017, the age-specific incidence rates of cervical cancer in Vietnam, measured per 100,000 women, were analyzed using data from the Institute for Health Metrics and Evaluation's Global Burden of Disease Study 2017 These rates were age-adjusted based on the Vietnamese female population from 1999 and 2017, highlighting significant trends in cervical cancer incidence over the years.

The cost of cervical cancer treatment varies globally, often proving to be expensive, particularly when late detection diminishes survival rates The 5-year survival rate for cervical cancer ranges from under 40% to over 70%, heavily influenced by the availability of HPV vaccination and screening programs Low- and middle-income countries (LMICs) experience significantly lower survival rates due to the challenges of early detection, which is crucial for curing invasive cervical cancer Without the implementation of effective intervention programs, cervical cancer is poised to remain a significant public health challenge in Vietnam for the foreseeable future.

Strategy to eliminate cervical cancer in the world and Viet Nam

Global studies confirm the HPV vaccine's effectiveness for girls and cervical screening for women in preventing cervical cancer The WHO recommends three key actions for eradication: vaccinating girls before age 15, screening women with a high-performance test by ages 35 and 45, and treating women with precancerous conditions and invasive cancer A systematic review of high-income countries, including the EU, USA, Canada, and Australia, highlights the integration of these strategies in national health policies, aiming for fewer than four cervical cancer cases per 100,000 women Prevention programs emphasize HPV vaccination for girls prior to sexual activity and cervical cancer screening for women aged 25 and older, including those previously immunized.

In the US, cervical cancer screening recommendations vary among institutions, but generally, women aged 21 to 65 should undergo screening at intervals of 3 to 5 years, depending on the testing method Specifically, women aged 21 to 29 should be screened every three years using the cytology method, while those aged 30 to 65 are advised to have co-testing every five years Additionally, co-testing can be initiated for women starting at age 25.

(27) Besides common screening methods, the US Preventive Services Task Force

2012 recommended co-testing as a primary method for the 30-65 women age group

In Europe, cervical cytology and HPV testing are the primary methods for national screening strategies, with some countries adopting HPV testing as the main screening approach in line with European guidelines The target demographic includes women aged 30 to 65 years In Australia, the National Cervical Screening Program updated its screening guidelines in 2017, transitioning from biennial cytology for women aged 20 to 65 to a five-year HPV test for women aged 25 to 74.

Low- and middle-income countries (LMICs) face significant challenges in implementing cervical screening programs due to socioeconomic factors and limited resources, despite cervical cancer being a leading cause of death among women High-income nations are more capable of providing HPV vaccination programs for pre-adolescent girls, primarily because of the prohibitive costs associated with the HPV vaccine Without international financial support, LMICs struggle to meet their vaccination needs In Vietnam, research by Hoang et al assessed the cost-effectiveness of HPV vaccines, Cervarix and Gardasil, for pre-adolescent girls, finding that vaccination would only be cost-effective if the price could be negotiated to $4.55 per dose, which is significantly lower than the current market price.

In 2017, the cost per Disability-Adjusted Life Year (DALY) averted was $8,000, indicating that cervical cancer screening strategies in low- and middle-income countries (LMIC) are highly cost-effective Literature reviews show that, in the absence of HPV vaccination, one-time cervical screening significantly reduces the prevalence of cervical cancer in women aged 35 and older Scientific studies confirm that early cervical cancer screening can lower morbidity and mortality rates for decades to come However, the search for an effective screening method in LMIC countries continues due to inadequate facilities and limited resources.

In Vietnam, the Ministry of Health (MOH) approved a national action plan for cervical cancer prevention and control for 2016-2025 and established guidelines in 2019 The plan is funded through international aid, as cervical screening and vaccination for girls are not included in the national health insurance program The guidelines outline six screening pathways for women aged 21-65, including cervical cytology, HPV testing, and colposcopy, with co-testing allowing for extended screening intervals of 5 years compared to the 2- or 3-year intervals of other methods However, the lack of public funding means that out-of-pocket payments may restrict access to cervical cancer screening.

Since its introduction in Vietnam in 2009, the HPV vaccine has not been included in the national vaccination program With support from the Bill & Melinda Gates Foundation, 6,358 girls received three doses of the vaccine in two provinces Currently, there is no available data on the number of girls who have voluntarily chosen to receive the HPV vaccine.

The HPV vaccine is priced affordably at approximately $46; however, the total cost for the recommended three doses can exceed $138 Furthermore, many women of childbearing age show a low willingness to pay for the HPV vaccine for their children upon learning its actual cost.

Figure 1.5: the proportion of nonvaccinated women willing to receive the HPV vaccine before and after knowing its price

As of now, there are no published reports on the national action plan for cervical cancer prevention and control, and data on the percentage of Vietnamese women undergoing cervical cancer screening or their preferred screening methods remains unclear The existing evidence regarding current cervical cancer screening practices among Vietnamese women is limited The national action plan aims to achieve a screening rate of 60% among women aged 30 to 54 by 2025, suggesting that currently, only about 30% of the targeted population is being screened.

In 2021, the HPV test generally costs more than colposcopy and cervical cytology As a result, women may opt for cervical cytology for screening when paying out of pocket, given colposcopy's lower effectiveness in early cervical cancer detection and its more affordable price compared to the HPV test.

Cervical screening methods

Cervical cancer screening primarily utilizes cervical cytology, colposcopy, and HPV testing, often in combination, with HPV testing frequently paired with cervical cytology (co-testing) Cervical cytology, the earliest screening method, involves two specimen preparation techniques: conventional smear (Pap smear) and liquid-based cytology (LBC), interpreted according to the 2014 Bethesda system Colposcopy, which employs acetic acid or Lugol’s iodine for cervix inspection, is categorized as negative or positive but is not recommended as a primary screening method due to its low effectiveness in detecting early HPV infection symptoms In contrast, HPV testing identifies high-risk HPV types in women through cell samples Screening strategies for cervical cancer vary by country, reflecting different levels of affordability and access to healthcare.

Cervical cytology has long been considered a primary global screening method, but with inadequate healthcare facilities, colposcopy has emerged as a viable alternative Liquid-based cytology (LBC) represents the next generation of cervical screening, demonstrating significantly higher efficacy in detecting squamous cell carcinoma recurrence compared to traditional Pap smears Over the past two decades, the importance of HPV testing has become clear, leading to its adoption as a replacement for cytology in screening and post-treatment follow-up, which has also increased HPV vaccination coverage However, as vaccination reduces lesion prevalence, the sensitivity and positive predictive value of cytology may decline Unlike morphological assessments, HPV tests focus on identifying genomic types to pinpoint at-risk patients for cervical cancer Nonetheless, existing HPV tests can produce a considerable number of false positives, potentially leading to unnecessary treatments for healthy women, particularly in the absence of cytology triage Recent studies indicate that co-testing is a more cost-effective approach than using cervical cytology or HPV testing alone, due to fewer screening cycles and improved cervical cancer detection rates.

Target age group and screening interval

Cervical cancer screening age recommendations vary by country, with high-income nations typically advising screenings for women aged 20 or 25 to 60 or 65 years The screening frequency is generally every two to three years with cytology and every five years for HPV testing or co-testing The World Health Organization (WHO) suggests that women in low- and middle-income countries (LMIC) should undergo cervical cancer screenings at least twice by the age of 35.

The World Health Organization recommends initiating cervical cancer screening programs for women aged 30 to 60, starting five years prior to the age when cervical cancer incidence begins to rise This approach aligns with the 45-year coverage of the HPV vaccine and emphasizes the importance of effective patient management and treatment strategies.

Women in high-income countries can undergo cervical cancer screening every three to five years as per national programs, while low- and middle-income countries (LMIC) face significant financial and resource constraints that limit such strategies The World Health Organization recommends that women aged 35-54 in LMIC with a sufficient screening history and no abnormal results may discontinue screening due to their low risk of HPV infection An adequate screening history is characterized by three consecutive negative cytology results or two negative co-testing results within the last ten years, with the latest test conducted within the past five years.

In addition, there is limited evidence about the optimal age when to stop screening

The appropriate age for cervical screening in each country should be determined by the specific incidence of cervical cancer, taking into account the experiences of other nations that have implemented similar screening programs.

A 2016 National Survey on Reproductive and Sexual Health revealed that Vietnamese adolescents engage in their first sexual intercourse at an average age of 18.7 years, with a notable lack of knowledge about safe sex practices Additionally, research by Tran Kim Ngoc et al indicates that the incidence of cervical cancer in Vietnam has been increasing, particularly among women aged 30-35, peaking in the 55 age group between 1999 and 2017.

The Ministry of Health (MOH) guidelines recommend cervical cancer screening for women aged 21 to 65, prioritizing those at risk between 30 to 50 years old This research proposes to initiate screening for cervical cancer at age 25 and continue until age 55 to facilitate early diagnosis and treatment, thereby preventing disease progression Co-testing, which combines cervical cytology and HPV testing, has proven to be more cost-effective than using either method alone, allowing for extended screening intervals from three years to five years Following the recommendations from the MOH and WHO, the researcher aims to develop a study involving 1,000 women aged 25 to 55, conducting three screenings over a 30-year period to assess the cost-effectiveness of the co-testing method.

Cost-effectiveness analysis of co-testing for cervical cancer screening

A literature review of 82 articles published in the PubMed database from 2011 to December 2021 reveals significant insights into the efficacy and effectiveness of co-testing methods for cervical cancer detection Among these, 13 studies assessed the cost-effectiveness of co-testing compared to alternative interventions, with six employing the Markov model for their evaluations However, the use of lifetime cycles in these models was not clearly stated, and only one study reported using monthly cycles, suggesting that most researchers likely utilized 1-year cycles In addition to the Markov model, various methodologies were identified, including the “Policy1-Cervix” model, decision trees, the FOCAL trial, retrospective cohort studies, and budget impact analyses, with research contributions from countries such as China, Thailand, Canada, the USA, and Germany.

The epidemiological development of cervical cancer encompasses seven key health states: (i) HPV negative (HPV -), (ii) high-risk HPV infection (HPV hr+), (iii) mild cervical intraepithelial neoplasia (CIN 1), (iv) moderate cervical intraepithelial neoplasia (CIN 2), (v) cervical intraepithelial neoplasia or carcinoma in situ (CIN 3), (vi) cervical cancer (CC), and (vii) death.

In Vietnam, research on the cost-effectiveness of cervical cytology is limited, with only two studies conducted: Suba et al in 2001 and Kim et al in 2008 Suba et al identified cervical cytology as the optimal screening method at the time, primarily comparing pap smears to no screening, which may not be relevant for current policymakers Meanwhile, Kim et al focused on the cost-effectiveness of HPV vaccination combined with cervical screening every five years, finding it cost-effective if the vaccination cost per girl was below $25 There is a notable lack of studies evaluating the cost-effectiveness of newer screening methods, such as co-testing, in Vietnam Comparing the efficiency of co-testing with cervical cytology will provide valuable insights for determining the most effective cervical cancer screening strategy in the country.

Research conceptual framework

In 2019, Vietnam introduced national guidelines for cervical cancer prevention and control, offering six screening options for women aged 21 to 65 However, the absence of a recommended primary screening method that is cost-effective may lead to inefficient resource use Given the scarcity of healthcare resources, prioritizing an efficient screening approach is essential This principle underpins health economics, prompting the researcher to conduct a cost-effectiveness analysis of cervical cancer screening specifically for women aged 25 to 55 in Vietnam The study's conceptual framework is illustrated in the accompanying graph.

Figure 1.6: The conceptual framework of the study

+ Quality adjusted life years + Mortality

HPV (-), HPV hr(+), CIN1/2/3, Cervical cancer

+ Literature review + Secondary data + Expert consultation

Research methodology

Research methodology for the objective 1

Research articles or reports that satisfied the following criteria were selected for literature review:

- Types of studies: CEA study

- Participants: Women aged 25 to 55 years old

- Full text report written by English and Vietnamese

This article reviews research published between 2011 and December 2021, utilizing bibliographic databases such as the Cochrane Central Register of Controlled Trials (CENTRAL), the Cochrane Gynecological Cancer Collaborative Review Group (CGCG) trial register, PubMed, and Web of Science Additionally, the researcher examined unpublished or grey literature, adhering to specific inclusion criteria, by accessing doctoral and master’s theses from libraries in Sweden (available at https://www.diva-portal.org) and Hanoi University of Public Health, Vietnam (accessible at http://opac.huph.edu.vn/opac/).

A comprehensive search in bibliographic databases was conducted using keywords related to cost-effectiveness analysis (CEA) and cervical cancer, including terms like "cervical intraepithelial neoplasia" (CIN) and screening methods After eliminating duplicates, the selected studies were assessed for quality based on the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) Eligible papers were then compiled into a summary table, detailing the publication year, methodology, model, objectives, target population, perspective, willingness to pay, cycle and time horizon, discount rates, interventions, comparators, sources of input parameters, and key results.

Research methodology for the objective 2

- Cost – effectiveness analysis: May 2022 – July 2022

As mentioned in the previous part, the incidence of cervical cancer in Viet Nam increased rapidly by age from women 30 to 34 years old and peaked in the 55 –

59 age group before decreasing gradually in older age groups The development of

Cervical cancer (CC) typically develops after a prolonged period following HPV infection, and with the average age of first sexual initiation among Vietnamese youth being 18.7 years, women aged 25 to 55 are identified as the primary focus of this research This demographic is particularly vulnerable to cervical cancer due to insufficient HPV vaccination coverage and the absence of national health insurance reimbursement in Vietnam.

The co-testing method has been shown to be more cost-effective than traditional cytology and HPV tests for routine cervical cancer (CC) screening, particularly for women up to a certain age limit However, low- and middle-income countries (LMICs) face challenges in implementing routine CC screening due to resource constraints The World Health Organization (WHO) recommends a minimum of three consecutive CC screenings as an effective strategy for women in LMICs This study aims to evaluate the cost-effectiveness of the co-testing method in Vietnam, comparing it to the cytology method, which has been recognized for its cost-effectiveness for over a decade Utilizing a Markov model, the research will analyze cervical cancer screening strategies for Vietnamese women aged 25 to 55, specifically contrasting the co-testing approach (three screenings at five-year intervals) with the cytology method (five screenings at two-year intervals).

- The 25-29 group was screen at the age of 25, 30 and 35 by the co-testing method, and at the age of 25, 27, 29, 31, 33 by the cytology method

- The 30-34 group was screen at the age of 30, 35 and 40 by the co-testing method, and at the age of 30, 32, 34, 36, 37 by the cytology method

In the CEA study discussed, the Markov model was utilized for evaluating co-testing methods, grounded in the natural epidemiology of cervical cancer (CC) development through a comprehensive review of medical literature and findings from randomized control trials and retrospective cohort studies The researcher also consulted various natural history models of high-risk HPV infections leading to oncogenesis and sought input from clinical experts for necessary adjustments While other models shared similar health states, four clinical experts did not raise any concerns regarding the proposed Markov model but recommended incorporating information about ASC-US, HSIL, and LSIL results from cytology tests Following the Ministry of Health's treatment guidelines, women with ASC-US cytology results were addressed accordingly.

The United States follows a similar follow-up pathway for HPV-positive women, where those with LSIL and HSIL results are treated as CIN 1 and CIN 2/3, respectively Therefore, no adjustments to the Markov model were required This model is based on the work of Felix, J C et al.

In 2016, a simulation was conducted to model the natural progression of HPV infection over one-year cycles, focusing on the cost-effectiveness of cervical screening using the co-testing method in comparison to the traditional cytology method.

The Markov model operates under several key assumptions: it analyzes 1,000 undiagnosed women aged 25-55 who are engaging in sexual behavior, all of whom do not have cervical intraepithelial neoplasia (CIN) or cervical cancer (CC) and will undergo cervical cancer screening through co-testing and cytology Additionally, none of the participants have received HPV vaccinations or undergone hysterectomy, and all are fully engaged in screening programs Importantly, every identified case, including HPV-positive individuals and those with CIN1, CIN2, CIN3, and CC, will receive appropriate treatment The model utilizes Quality-Adjusted Life Year (QALY) weights to evaluate health outcomes, structured around the natural history of HPV infection.

The progression of cervical health can be categorized into several states: State A represents HPV negative (HPV -) status, while State B indicates a high-risk HPV infection (HPV +) Moving to State C, mild cervical intraepithelial neoplasia (CIN 1) is present, which can progress to State D, characterized by moderate cervical intraepithelial neoplasia (CIN 2) Further progression leads to State E, where cervical intraepithelial neoplasia or carcinoma in situ (CIN 3) is diagnosed, ultimately culminating in State F, which denotes the presence of cervical cancer (CC) The cycle concludes with State G, representing death, typically occurring within a year of diagnosis.

Figure 2.1: Markov model of Felix, J C et al (2016)

This study assesses the cost-effectiveness of cervical cancer (CC) screening for Vietnamese women aged 25 to 55 by comparing co-testing with the cytology method over a 30-year period The target group will undergo co-testing three times at five-year intervals, while the alternative approach involves five screenings every two years A Microsoft Excel template is utilized for simulation and to calculate the incremental cost-effectiveness ratio (ICER).

Following the initial screening, women were categorized into four conditions based on test performance, identifying true/false negatives and true/false positives for HPV cases They were then divided into two subgroups: detected and undetected groups As noted in the literature review, women with HPV infection and precancerous conditions (CIN1/2/3) often experience recovery and significant clearance of their HPV infection post-treatment Consequently, undetected women across all health states continued to follow the natural progression of their condition.

Death from CC and other causes*

Patients with cervical cancer can progress to all-cause death from various health states, ranging from state A to state F Women diagnosed through cervical cancer screening receive treatment and follow standard progression and regression transition probabilities According to Ministry of Health guidelines, false positive HPV cases undergo triage tests twice within a year: those screened using the co-testing method will be re-screened with the same method, while those screened by cytology will be tested again with an HPV test.

2.2.5 Time horizon, cycle length and analytical perspective:

The study involved a cohort of 1,000 women aged 25, tracking them until they reached the maximum screening age, with assessments conducted annually Due to time constraints, the researcher utilized a provider perspective for the cost-effectiveness analysis.

After adjusting all costs to 2021 inflation rates and applying a discounting rate of 3% for Vietnam, an Incremental Cost-Effectiveness Ratio (ICER) was calculated to assess the efficiency of the co-testing method versus the cytology method The ICER represents the additional costs incurred by adopting the co-testing method in relation to the Quality-Adjusted Life Years (QALYs) gained A lower ICER indicates a more cost-effective approach, and according to WHO guidelines, an ICER below 1 to 3 times the GDP per capita is considered cost-effective In 2020, Vietnam's GDP per capita was reported at $2,785.70 by the World Bank.

The conversion rate for Vietnamese Dong (VND) to US Dollar (USD) is set at 22,860 VND per USD, according to Vietcombank This translates to a GDP per person of 63.68 million VND, equivalent to approximately 2,786 USD, and a GDP per person of 191.04 million VND, which is about 8,357 USD for three GDPs.

2.2.7 Input parameter and data sources:

The input parameters identified through a literature review and secondary data were validated and refined through consultations with experts The researcher conducted interviews with a cancer specialist, reproductive health professionals, and an additional expert in the field to ensure comprehensive insights.

The Incremental Cost-Effectiveness Ratio (ICER) plays a crucial role in financing national health insurance, particularly in the context of cervical cancer and reproductive health Specialists in these fields contributed valuable insights into the disease's development, patients' Quality-Adjusted Life Year (QALY) weights across various health states, and the transition probabilities between these states They also evaluated the specificity and sensitivity of different cervical cancer screening methods and treatment scenarios through a Markov model Additionally, financial experts collaborated with researchers to ensure treatment costs were aligned with the latest Ministry of Health (MOH) regulations.

Ethical issues

The research proposal received approval from the Ethical Committee of Hanoi University of Public Health Participants were informed about the study's objectives and purpose, ensuring their right to consent or decline participation The study utilized existing databases without clinical intervention, preserving the customs and habits of participants No personally identifiable information was used or shared in the research report All participant information was collected and stored in compliance with confidentiality, strictly for research purposes.

2022, the ethical issues of the study were approved and accepted by the Ethical Committee of Hanoi University of Public health following the letter no 154/2022/YTCC-HD3 (Annex 8).

Results

The literature review of cost-effectiveness studies of cervical cancer screening for

screening for women by the co-testing method

A comprehensive search of bibliographic databases, including The Cochrane Central Register of Controlled Trials, The Cochrane Gynecological Cancer Collaborative Review Group trial register, PubMed, and Web of Science, from 2011 to December 2021, identified 82 articles on the efficacy and cost-effectiveness of the co-testing method for cervical cancer detection After applying selection criteria, 13 studies were included in the literature review, with 11 focusing on cost-effectiveness analysis within cervical screening programs Additionally, two studies on budget impact and cost-effectiveness analysis of the co-testing method were incorporated Other studies, being observational or randomized control trials without cost-effectiveness analysis, were excluded Due to the lack of studies targeting the specific screening age of 25-55, those with broader age intervals were selected A summary of the selected studies is provided in Table 1, with further details in Annex 5.

Figure 3.1 PRISMA diagram for literature reviews

Thirteen studies across various countries, including the Netherlands, USA, Ireland, Korea, Germany, Greece, British Columbia, Canada, Australia, and China, were analyzed (Table 3.1) Among these, seven articles utilized cost-effectiveness analysis (CEA) with Markov modeling, while the remaining six employed diverse methodologies such as the “Policy1-Cervix” platform, the MISCAN-Cervix model from the Cancer Intervention and Surveillance Modeling Network, decision tree modeling, prospective trials, retrospective cohort studies, and a combination of Markov models and decision trees Although some researchers did not clearly define their costing perspective in relation to cervical screening algorithms, it is generally assumed that most studies adopted the payer perspective for measurement Only two studies incorporated a societal perspective in their cost analysis.

12) Records removed due to unable access to full text (n = 20)

Not written in English (n = 2) Randomized control trial studies (n = 5)

Identification of studies via databases

Ident if ica ti o n Scre en ing Incl ud ed

Records excluded due to not analysis cost-effectiveness of co- testing (n = 32)

Reports not retrieved due to analysis of clinical benefits (n = 27)

To conclude, a majority of records applied the Markov model for their decision model under the provider perspective to inform recommendations for policy makers

Table 3.1: List of selected studies

No Article Authors Published year

1 Cost- effectiveness analysis of primary human papillomavirus testing in cervical cancer screening:

Results from the HPV FOCAL Trial

2 Health economic evaluation of primary human papillomavirus screening in urban populations in

3 Economic evaluation of HPV DNA test as primary screening method for cervical cancer: A health policy discussion in Greece

4 Screening capacity and cost-effectiveness of the human papillomavirus test versus cervicography as an adjunctive test to Pap cytology to detect high- grade cervical dysplasia

5 Primary HPV testing versus cytology-based cervical screening in women in Australia vaccinated for HPV and unvaccinated: effectiveness and economic assessment for the

Jie-Bin Lew et al,

6 The budget impact of cervical cancer screening using HPV primary screening

7 A model to evaluate the costs and clinical effectiveness of human papilloma virus screening compared with annual Papanicolaou cytology in

Karl Ulrich Petry et al,

8 Optimal Cervical Cancer Screening in Women

9 The Clinical and Economic Benefits of Co-Testing

Versus Primary HPV Testing for Cervical Cancer

10 Cervical Cancer Screening in Partly HPV

11 Cost Effectiveness of Human Papillomavirus-

16/18 Genotyping in Cervical Cancer Screening

12 Long-term costs of introducing HPV-DNA post- treatment surveillance to national cervical cancer screening in Ireland

13 Screening for Cervical Cancer: A Modeling Study for the US Preventive Services Task Force

Five out of thirteen studies on cervical cancer screening strategies were conducted in the United States, primarily modeling a theoretical cohort of women aged 25 or 30 over a 40-year period, regardless of HPV vaccination status Notably, Shalini L Kulasingam and colleagues utilized a Markov model that tracked women from ages 12 to 100, aligning with the US Preventive Services Task Force recommendations In contrast, Jane J Kim and her team specifically examined screening strategies for vaccinated women starting at ages 25 or 30 Other research from countries such as the Netherlands, Ireland, Germany, Greece, and British Columbia, Canada, similarly focused on cohorts aged 25 to 65.

Research from Australia and China reveals notable differences in HPV screening studies While the Australian study examined participants aged 10 to 84, the Chinese research focused solely on unvaccinated women, as there is no HPV vaccination program in the country The age range for screening varied across studies, typically starting at 25 and ending at 65 years However, some studies suggest that screening women over 65 may not provide significant benefits.

Table 3.2: Cervical cancer screening age interval

No Article Country Starting age

Results from the HPV FOCAL Trial

2 Health economic evaluation of primary human papillomavirus screening in urban populations in

3 Economic evaluation of HPV DNA test as primary screening method for cervical cancer: A health policy discussion in Greece

4 Screening capacity and cost-effectiveness of the human papillomavirus test versus cervicography as an adjunctive test to Pap cytology to detect high- grade cervical dysplasia

5 Primary HPV testing versus cytology-based cervical screening in women in Australia vaccinated for HPV and unvaccinated:

No Article Country Starting age

Finished age effectiveness and economic assessment for the

6 The budget impact of cervical cancer screening using HPV primary screening

7 A model to evaluate the costs and clinical effectiveness of human papilloma virus screening compared with annual Papanicolaou cytology in

8 Optimal Cervical Cancer Screening in Women

9 The Clinical and Economic Benefits of Co-Testing

Versus Primary HPV Testing for Cervical Cancer

10 Cervical Cancer Screening in Partly HPV

11 Cost Effectiveness of Human Papillomavirus-

16/18 Genotyping in Cervical Cancer Screening

12 Long-term costs of introducing HPV-DNA post- treatment surveillance to national cervical cancer screening in Ireland

13 Screening for Cervical Cancer: A Modeling Study for the US Preventive Services Task Force

Authors developed diverse screening scenarios based on age, focusing on initiation and completion of screenings, transitions to new tests, and routine screening intervals aligned with national guidelines Typically, HPV-based and co-testing strategies are compared to cytology-based methods However, a study by Adam Keane et al used no screening scenario as a comparator, while Jane J Kim et al suggested a specific threshold for evaluation.

$50,000 to $200,000 per QALY gained as reference range for cost-effectiveness assessment

Countries have developed various cervical cancer (CC) screening strategies, utilizing either cytology tests, HPV tests, or a combination of both, along with additional tests These strategies vary significantly across nations, particularly in the selection of front-line screening tests and the triage methods employed for positive results Overall, research typically evaluates the cost-effectiveness of eight distinct screening pathways.

CC screening: (i) primary cytology (pap or LBC) triage with HPV test or colposcopy;

(ii) primary cytology triage with co-testing or HPV test; (iii) primary HPV test triage with cytology or colposcopy; (iv) primary HPV genotyping test (with or without HPV

The article discusses various triage methods for cervical cancer screening, including cytology and colposcopy, as well as primary co-testing with and without genotyping Key strategies include the use of HPV testing, cervicography via the TeleCervico system, and the application of p16/Ki-67 dual-stained cytology as a biomarker to identify high-grade CIN and cervical cancer Research by Karl Ulrich Petry et al highlights the effectiveness of this biomarker, while Steffie K Naber et al recommend a three-time consecutive screening approach for unvaccinated women aged 35-59 to evaluate cost-effectiveness Screening frequencies varied across studies, ranging from one to twelve times based on the screening strategy employed Detailed comparisons of interventions are available in table 3.3.

Screening test Interval Screening test

Primary HPV testing with reflex LBC (co-testing)

2 Health economic evaluation of primary human papillomavirus screening in urban populations in China

(i) primary cytology (pap or LBC) triage with colposcopy

(ii) Primary HPV DNA test with cytology triage

(iii) primary HPV genotyping test (with HPV 16/18) triage with colposcopy

(iv) primary co-testing triage with colposcopy

(v) Primary HPV DNA test with HPV 16/18 genotyping, and cytology triage

(vi) Primary cytology test with HPV DNA triage

(vii) Colposcopy and biopsy management

HPV DNA test as primary screening method for cervical cancer: A health policy discussion in

Cytology (current practice) 1 year or 3 years

Cytology 1 year the HPV test with simultaneous HPV 16 and

3 years or 5 years the HPV test with reflex 16 and 18 genotyping

3 years or 5 years the HPV test with no genotyping

Co-testing with cytology and the HPV test with simultaneous 16 and 18 genotyping

Co-testing with cytology and the HPV test with reflex

Co-testing with cytology and the HPV test with no genotyping

4 Screening capacity and cost-effectiveness of the human papillomavirus test versus cervicography as an adjunctive test to Pap cytology to detect high- grade cervical dysplasia cervicography (TeleCervico system) + pap

Not mention HPV test genotyping

5 Primary HPV testing versus cytology-based cervical screening in women in Australia vaccinated for HPV and unvaccinated: primary cytology (pap) triage with colposcopy

3-yearly intervals for ages 25-49 years and 5- yearly intervals for conventional cytology, no HPV triage testing

Interval effectiveness and economic assessment for the National Cervical

Screening Program ages 50-64 years primary cytology (LBC) triage with colposcopy

3-yearly intervals for ages 25-49 years and 5- yearly intervals for ages 50-64 years

Primary HPV genotyping test triage with cytology and colposcopy with cytology triage of all oncogenic HPV positive women

5 years primary co-testing triage with colposcopy

6 The budget impact of cervical cancer screening using HPV primary screening primary HPV genotyping test with HPV 16, 18 and 12 hr HPV triage with cytology or colposcopy

3 or 5 years primary cytology (pap or LBC) triage with HPV testing or colposcopy

3 years primary co- testing without genotyping triage with colposcopy or co-testing again

Co-testing with genotyping triage with colposcopy or co-testing again

7 A model to evaluate the costs and clinical a positive HPV test followed by Pap cytology

Interval effectiveness of human papilloma virus screening compared with annual

In Germany, a positive HPV test is typically followed by dual-stained cytology for p16/Ki-67 If the HPV test indicates HPV-16 or HPV-18, a colposcopy is recommended For other HPV subtypes, dual-stained cytology is utilized Additionally, co-testing with HPV and Pap tests is an essential part of the screening process.

Good value for money according to thresholds of $50 000 to $200 000 per quality-adjusted life- year (QALY) gained Women vaccinated with

Cytology when 21 and switch to HPV test when 30

3 years for cytology and 5 years for HPV test

Cytology when 21 and switch to cotest test when 30

3 years for cytology and 5 years for cotest

HPV test 3 years, 4 years, 5 years,

3 years for cytology and 3-5 years for HPV test

3 years for cytology and 5 years for cotest

Co-testing: Pap (ThinPrep Pap Test and Aptima HPV Assay, Hologic, Inc.) plus HPV mRNA testing with genotyping for HPV 16/18

HPV DNA testing with genotyping for HPV 16/18 and reflex cytology (cobas®

HPV test, Roche Di- agnostics US) (HPV primary)

10 Cervical Cancer Screening in Partly HPV Vaccinated

(A) primary HPV screening with reflex cytology triage and cytology triage after six months (future Dutch screening program),

(B) primary cytology with reflex HPV triage,

(C) combined primary HPV and cytology (i.e co- testing) with HPV triage after 12 months,

The study evaluates the cost-effectiveness of primary cytology combined with HPV triage for unvaccinated women, analyzing screening intervals ranging from 5 to 20 years and lifetime screenings between 1 and 12 It calculates the Incremental Cost-Effectiveness Ratio (ICER) of a screening strategy tailored for the pre-vaccination population in comparison to one optimized for vaccinated women.

Interval cytology triage after 18 months (current Dutch screening program)

(1) primary cytology (pap or LBC) triage with HPV testing or colposcopy/ cytology with reflex HPV testing for atypical squamous cells of undetermined significance (ASC-US);

(2) primary co-testing without genotyping) triage with colposcopy or co- testing again/co-testing with cytology and HPV testing;

(3) primary HPV test triage with cytology pap /HPV with reflex to cytology

(4) primary HPV genotyping test (with HPV 16/18) triage with cytology or colposcopy/HPV with 16/18 genotyping and reflex cytology (ASC-US threshold)

12 Long-term costs of introducing HPV-DNA post-treatment surveillance to national cervical cancer screening in Ireland co-testing (cytology plus human papillomavirus DNA testing)

2 semi-annual Cytology at 6 months post- treatmen t, then at

18 months and annually , for 5-10 years thereafte r

Cancer: A Modeling Study for the US Preventive

Co-testing: human papillomavirus DNA testing in conjunction with cytology

The costing perspective in various studies primarily focused on healthcare service payers, with a social perspective identified in two studies by Steffie K Naber et al and Jane J Kim et al In contrast, Juan C Felix et al and Ian Cromwell et al only assessed direct medical costs associated with service delivery To evaluate screening test costs, researchers utilized fee schedules sanctioned by national medical authorities, while treatment costs were derived from published studies and reviews by at least two clinical experts, establishing average treatment and follow-up costs for cervical cancer across all states Notably, Anastasios Skroumpelos et al faced data limitations and adapted Spanish data for their Greek analysis Additionally, authors employed a positive discount rate of 3% to 5% for both QALYs and costs, although the rationale for these rates was not provided in their findings.

Authors calibrated transition probabilities and QALY weights using extensive population trials and clinical data Key studies in the US, particularly the ATHENA trial, provided epidemiological estimates for unvaccinated women in the screening population Jane J Kim et al analyzed optimal screening pathways for vaccinated women using data from the New Mexico HPV Pap Registry and the National Health and Nutrition Examination Survey In the Netherlands, researchers validated data from the Costa Rica Vaccine Trial and the PATRICIA trial, which evaluated the HPV vaccine's clinical benefits for women who were HPV naïve at enrollment These studies enabled an analysis of the cost-effectiveness of screening among pre-vaccinated, vaccinated, and unvaccinated women within their simulation cohorts.

The Multicentric Cervical Screening study (HERMES) utilized clinical, epidemiological, and diagnostic test performance data in Greece, although the target population remains unspecified due to inaccessible trial details In contrast, British Columbia and Canada conducted the HPV-FOCAL randomized controlled trial, focusing on Human Papillomavirus testing for cervical cancer screening Both the ATHENA and HPV-FOCAL trials examined HPV testing as a primary screening method in the US, with ATHENA specifying the Cobas HPV Test for performance evaluation In Ireland, data from CervicalCheck, the National Cervical Screening Programme initiated in 2008, was analyzed to assess the clinical and economic advantages of co-testing in post-treatment cervical cancer surveillance Meanwhile, studies in Australia and China gathered data from various healthcare facilities, with Australia sourcing information nationwide and China focusing on Shenzhen and surrounding cities in the Pearl River Delta of Guangdong province.

The studies showed that co-testing proved its cost-effectiveness in preventing

Recent studies indicate that cervical cancer (CC) incidence and mortality rates are lower with co-testing compared to cytology methods, although differences are not significant when HPV testing is used as a benchmark Co-testing not only resulted in the lowest CC incidence but also excelled in early detection, leading to a higher number of Quality-Adjusted Life Years (QALYs) gained, despite the lack of statistical significance However, it also incurred the highest cost per screened woman, influencing policy decisions Consequently, many studies favor HPV-based screening strategies over co-testing due to cost reductions When comparing cytology-based screening methods, such as Pap tests and liquid-based cytology (LBC), primary HPV testing emerged as more cost-effective, yielding fewer deaths from missed diagnoses Additionally, this approach led to reduced costs, extended screening intervals, and improved QALYs Research by Steffie K Naber et al suggests that screening intervals could be safely extended to six years for pre-vaccinated cohorts and twelve years for vaccinated cohorts, as opposed to the previously recommended five years Despite the superiority of primary HPV testing over cytology, it notably increased referral rates for colposcopy per prevented death, particularly among unvaccinated women when herd immunity reached 100%.

Parameter inputs for CEA of the co-testing method

The ATHENA study provides valuable insights into the prevalence of high-risk HPV types within the general population, particularly in contexts similar to Vietnam, where HPV vaccination coverage is notably low at just 2%.

(106) For progression and regression transition probabilities, the data from the study

The article "Screening for Cervical Cancer: A Decision Analysis for the U.S Preventive Services Task Force" by Juan C Felix et al examines various screening methods for cervical cancer While studies conducted in Korea and China employed decision tree and Policy1-Cervix models to evaluate the cost-effectiveness of these methods, their findings are not applicable to my Markov model analysis.

Recent literature reviews indicate that most studies calculated transition probabilities based on epidemiological data from the past 15 years, primarily from Western countries, during which the HPV vaccine had not yet been developed This limitation hindered the understanding of the natural progression of cervical cancer (CC) Additionally, the data lacked comprehensive range values and transition probabilities from CIN2 to CIN3 To address these gaps, findings from the study "The predicted effect of changes in cervical screening practice in the UK: Results from a modelling study" were utilized.

“Effectiveness and cost-effectiveness of eliminating cervical cancer through a tailored optimal pathway: a modelling study” (109,110) were taken The transition probabilities were presented in the table 3.6

The data collected underwent validation through consultations with four clinical experts Due to the absence of available data in Vietnam, the experts were unable to offer suggestions for data adjustments Nonetheless, randomized controlled trials conducted in populations with low HPV vaccine coverage, published in the Cochrane Central Register of Controlled Trials, serve as suitable and reliable reference sources.

3.2.2 Effectiveness of co-testing and cytology method

After conducting a thorough search in bibliographic databases such as the Cochrane Database of Systematic Reviews, PubMed, and Web of Science, the researcher identified a total of 25 articles focused on the effectiveness of co-testing and 50 articles on the effectiveness of cytology Subsequently, four reports were selected based on their abstracts and summary information.

Figure 3.2: PRISMA diagram for test effectiveness

Identification of studies via databases

Ident if ica ti o n Scre en ing Incl ud ed

Records excluded due to not analysis test effectiveness (n 1 = 18

Recent analyses of co-testing performance in cervical cancer (CC) screening indicate that co-testing outperforms cytology alone in sensitivity but falls short in specificity Specifically, the pooled results from Tong Li et al show a sensitivity of 0.937 and specificity of 0.858, which are higher than the findings from Mamiko Onuki et al., with sensitivity at 0.92 and specificity at 0.76 The results from Tong Li et al are particularly significant for the cost-effectiveness analysis of this study, as they provide valuable diagnostic insights for CC screening, unlike the latter study, which focused on post-treatment data.

The effectiveness of the cytology method is supported by data from Alejandra Castanon et al., who focused on assessing its sensitivity In contrast, Teruhiko Terasawa et al highlighted varying ratios of sensitivity and specificity among different screening tests without reporting cytology performance in cervical cancer (CC) screening separately To supplement the analysis, the specificity data for cytology was sourced from Tong Li et al The findings regarding the effectiveness of cytology and co-testing are summarized in Table 3.5.

The systematic review reports did not clarify whether co-testing utilized the same or separate specimens, nor did they address the impact of different sampling processes on test effectiveness The Ministry of Health's guidelines also did not cover these aspects According to the National Cancer Institute, a single specimen can be used for HPV testing Additionally, a search of bibliographic databases revealed no records indicating that the sampling process for co-testing influenced test effectiveness, suggesting that specimen preparation may not be a significant concern in co-testing practices.

Clinical experts reviewed the collected data and provided recommendations, with one expert suggesting a decrease in the cytology test sensitivity to 0.7 for case analysis However, the researcher opted to utilize the original data, as the lower range of cytology sensitivity was 0.78, which closely aligned with the expert's opinion This data will be analyzed in a one-way deterministic sensitivity analysis (DSA).

Table 3.4: List of systematic review studies about co-testing and cytology effectiveness

No Article Authors Published year

1 Posttreatment human papillomavirus testing for residual or recurrent high-grade cervical intraepithelial neoplasia: a pooled analysis

2 Diagnostic value of combination of HPV testing and cytology as compared to isolated cytology in screening cervical cancer: A meta-analysis

1 Systematic Review and Meta-Analysis of Individual

Patient Data to Assess the Sensitivity of Cervical

Cytology for Diagnosis of Cervical Cancer in Low- and Middle-Income Countries

2 Comparative accuracy of cervical cancer screening strategies in healthy asymptomatic women: a systematic review and network meta‑analysis

Currently, there is a lack of Vietnamese data regarding the Quality-Adjusted Life Year (QALY) weights for various states in the Markov model Cost-effectiveness analyses of cervical cancer screening utilizing the co-testing method in countries like China, Taiwan, and Thailand have relied on utility weights derived from clinical trials conducted in other nations, including the USA and Canada.

The UK is set to calculate Incremental Cost-Effectiveness Ratios (ICERs), utilizing research data from Warner K H et al specifically for Vietnamese women The authors derived Quality-Adjusted Life Years (QALYs) for various cervical states, including CIN 1, CIN 2, CIN 3, and cervical cancer (CC), based on surveillance data from Canadian and US populations using the standard gamble method To prevent overestimation of utility for cervical cancer, the QALY for the CC state was averaged across treatment phases according to FIGO classifications Clinical experts did not provide recommendations for the proposed data, noting that HPV-positive patients with CIN 1/2/3 could significantly regain their health post-treatment, experiencing no disruptions in daily activities, including their sexual lives Therefore, the use of average QALY weights for CC patients is deemed reasonable, considering the diverse health conditions associated with different CC states.

3.2.4 Cost for co-testing method, cytology method and cervical cancer treatment from provider perspective

A study by Nguyen, A D et al estimated the medical resources required for cervical cancer treatment at central hospitals in Vietnam in 2017, considering the provider perspective The treatment costs for cervical cancer (CC) varied by severity, with similar pathways for CIN 1, CIN 2, and CIN 3, leading to the use of average costs for these scenarios To conduct a health economic evaluation, the 2017 costs were adjusted to 2021 using the Consumer Price Index from the Vietnam General Statistics Office Additionally, the costs for cytology and co-testing methods were sourced from price lists published by K Hospital and the National Hospital of Obstetrics and Gynecology.

The treatment services related to CC were reviewed by clinical experts, leading to a revised list that included published costs per service based on financial expert consultation Overall, the updated costs showed only slight differences compared to the findings from Nguyen, A D et al The researcher adjusted the treatment costs for analysis and examined a 20% fluctuation in costs in relation to changes in the ICER.

Method Base case Lower range

Upper range Distribution Source Sensitivity Co-testing 0.937 0.925 0.948 Beta Li T et al, (113)

Cytology 0.809 0.78 0.838 Beta Castanon A et al, (114)

Specificity Co-testing 0.858 0.855 0.860 Beta Li T et al, (113)

Cytology 0.951 0.949 0.953 Beta Li T et al, (113)

Upper range Distribution Source Well, HPV (-) or

CIN 1 0.97 0.87 1 Beta Huh W K et al, (115)

ICC 0.71 0.64 0.78 Beta Huh W K et al, (115)

Upper range Distribution Source Treatment of

Co-testing 73.91 $ Gamma Hospital price list

HPV hr 0,267 0,134 0,099 0,076 0,066 0,066 0,041 0,305 Beta Felix J C et al, Wright T C et al, (21,106) Annual transition probabilities

Progression for detected and undetected cases

HPV hr (-) to HPV hr (+) 0,05 0,01 0,01 0,01 0,01 0,005 0 0,17 Beta Felix J C et al, Kulasingam

S L et al, (21,116) HPV hr (+) to CIN1 0,054 0,054 0,054 0,054 0,054 0,054 0,0527 0,1121 Beta Felix J C et al, Wright T C et al, (21,106,109) HPV hr (+) to

0,006 0,006 0,006 0,006 0,006 0,006 0,0045 0,0075 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN1 to CIN2/CIN3 0,02 0,02 0,02 0,02 0,02 0,02 0,001 0,06 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN2/CIN3 to ICC 0,01 0,01 0,01 0,01 0,01 0,01 0,001 0,04 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN2 to CIN3 0,0389 0,0389 0,0797 0,0797 0,1062 0,1062 0,0389 0,1062 Beta Canfell K et al, (110)

HPV hr (+) to HPV hr (-) 0,5 0,25 0,25 0,15 0,15 0,05 0,05 0,5 Beta Felix J C et al, (21)

CIN1 to HPV hr (+)/(-) 0,1 0,1 0,06 0,06 0,06 0,06 0,058 0,27 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN2/CIN3 to CIN1 0,03 0,03 0,03 0,03 0,03 0,03 0,001 0,19748 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN2/CIN3 to HPV hr (-) 0,03 0,03 0,03 0,03 0,03 0,03 0,001 0,19748 Beta Felix J C et al, Kulasingam

S L et al, (21,116) CIN3 to CIN2 0,0135 0,0135 0,0135 0,0135 0,0135 0,0135 0,01013 0,01688 Beta Xia C et al, (109)

4 Please see annex 6 to explain how to use transition probabilities and distribution calculation

CEA results – co-testing 3 times v.s cytology 5 times

3.3.1 Incremental cost-effectiveness ratio (ICER)

The findings indicate that co-testing is not a cost-effective option compared to cytology in a simulated cohort of 1,000 women Specifically, across five age groups (25-29, 35-39, 40-44, 45-49, and 50-55), co-testing incurs higher costs and results in lower Quality-Adjusted Life Years (QALYs) For women aged 30-34, although the cost of co-testing is lower than that of cytology, transitioning to co-testing would lead to a decrease in QALYs While co-testing could be considered a cost-effective alternative if the savings per QALY lost exceeded the threshold, this is not applicable here, as the WHO recommended threshold is 2,786 USD per capita, while the savings amount to only 3 USD.

The cytology test is less effective in reducing the incidence of cervical cancer (CC) and cervical intraepithelial neoplasia (CIN 1/2/3) compared to co-testing, which can prevent 887 cases of CIN 1/2/3 and 32 cases of CC, while cytology only prevents 627 CIN cases and 24 CC cases across all age groups when compared to no screening However, co-testing results in nearly double the number of false positives compared to cytology Additionally, costs for older age groups are significantly lower, as the model only simulated the cohort until they reached 55 years old, resulting in fewer follow-up cycles and reduced expenses.

Table 3.9: Cost, QALY weights and ICER between alternatives (3% discount)

Co-testing 3 times Cytology 5 times Difference cytology v.s co-testing

3.3.2.1 One-Way Deterministic Sensitivity Analysis (DSA)

The top 15 factors significantly influencing the Incremental Cost-Effectiveness Ratio (ICER) across different age groups are presented in Annex 7 Furthermore, the researcher examined the impact of variations in the number of cervical cancer screenings using the cytology method on the ICER.

The analysis of the top 15 inputs affecting the Incremental Cost-Effectiveness Ratio (ICER) reveals significant differences across age groups The Tornado diagrams indicate that the ICER is most sensitive to the transition probability from HPV hr(-) to HPV hr(+) and the prevalence of HPV hr among women in all age categories While the cost and effectiveness of screening tests have minimal impact on the ICER in most age groups, they notably influence the 50-55 age group, where the cost of co-testing ranks third, and the HPV and cytology test costs rank sixth and seventh, respectively Treatment costs for cervical cancer (CC) significantly affect the ICER, especially in the 35-39 age group, where it ranks second, and in other groups (25-29, 30-34, and 40-44) where it ranks third However, this parameter has less influence on the ICER changes in the 50-55 age group.

A comparative analysis of the cost-effectiveness of three consecutive co-testing versus three consecutive cytology screenings for women aged 25-30 revealed that the co-testing method, with a cost of $1,202 and QALYs of 24.18, outperforms the cytology method, which costs $955 and yields 22.55 QALYs The incremental cost-effectiveness ratio (ICER) for co-testing stands at $152 per QALY gained, significantly below the GDP per capita threshold of $2,786 Additionally, 1,000 Monte Carlo simulations conducted under uncertain conditions confirmed that the frequency of screening has a substantial effect on the ICER, with results situated in the north-east quadrant.

The cost-effectiveness analysis of 1,000 Monte Carlo simulations and cost-effectiveness acceptability curves (CEAC) reveals significant insights into the incremental cost-effectiveness ratios, which fluctuate in both the north-west and south-west quadrants These findings indicate that the strategy of screening with co-testing three times is dominated by the comparator The CEACs further confirm that sensitivity analyses from probabilistic and deterministic approaches are robust, with a high probability of the incremental cost-effectiveness ratio (ICER) remaining below the willingness to pay (WTP) threshold in Vietnam, which is set between 1 to 3 GDP per capita However, the co-testing method shows a cost-effectiveness probability of only 40% to 50% across all age groups, declining to nearly 0% as the WTP threshold increases to $1,200, indicating it is not cost-effective at this level In contrast, the cytology method demonstrates a higher likelihood of cost-effectiveness, with probabilities ranging from 40% to 60% across all age groups, and it approaches 100% cost-effectiveness as the WTP threshold reaches $1,200, which is below 1 GDP ($2,786).

Women aged 25 to 29: CE plane and CEA Curve

Discussion

Main findings from the literature review

Screening age intervals for cervical cancer (CC) vary globally, with countries like Korea and Australia starting at 18 and continuing services up to 80 and 69, respectively In contrast, some nations initiate screening at 21 or 25, still offering services to older women However, studies indicate that the benefits of screening women under 21 do not outweigh the harms due to the high HPV clearance rate in this age group Additionally, screening for women over 65 is often less cost-effective, particularly for those aged 80 and above The frequency of screenings differs by national guidelines, but vaccinated women may extend their screening intervals to ten years, contributing to herd immunity Once herd immunity exceeds 50%, screening becomes cost-ineffective for unvaccinated women, suggesting a potential reduction in screening intensity for this group Some experts propose that women aged 25 to 65 may be eligible for discharge after two consecutive negative tests, while those 65 and older could be discharged after 15 years of negative results, supporting a revised screening routine.

This study proposes a cervical cancer screening age interval of 25 to 55, aligning with most studies and Ministry of Health recommendations While there is no consensus on the upper age limit for screening, it has been shown that screening older women is not cost-effective when considering potential harms The World Health Organization highlights the importance of screening at ages 35 and 45, reinforcing the relevance of the proposed upper age limit.

Cytology and HPV testing are essential for cervical cancer screening Cytology includes two specimen preparation methods: conventional smear (Pap smear) and liquid-based cytology (LBC) While LBC is considered a more advanced technique with a higher efficacy in detecting squamous cell carcinoma recurrence, recent US data indicates that the sensitivity and specificity differences between LBC and Pap smear are minimal A 2008 systematic review and meta-analysis also found that the Pap test does not exhibit lower sensitivity or specificity in detecting high-grade cervical intraepithelial neoplasia compared to LBC.

Many studies did not specify whether they used conventional Pap tests or liquid-based cytology (LBC) in their cervical cancer (CC) screening strategies, although most did indicate the type of HPV test employed The primary HPV tests detect the presence of low- and high-risk HPV types using molecular biology techniques, often referred to as HPV tests, HPV hybrid capture tests, or HPV DNA tests Advances in technology now allow for the genotyping of 37 HPV types, with various manufacturers producing these tests HPV assays are categorized into four groups: hybridization with signal amplification, PCR for ≥ 13 high-risk HPV genotypes, amplification of E6/E7 viral mRNA, and assays identifying HPV16 or HPV18 Authors often refer to these techniques generically as HPV genotyping tests or HPV PCR tests A few studies have assessed the cost-effectiveness of strategies using HPV 16/18 genotyping due to its significance in identifying high-risk cases of CC within the target population.

In 2003, the FDA approved co-testing, which combines cytology and HPV testing, as a standard cervical screening method for women aged 30 and older However, not all HPV testing platforms using liquid-based cytology (LBC) methods received FDA approval, necessitating validation studies for off-label applications The 2012 guidelines from the American Cancer Society and other organizations recommended co-testing as the primary screening approach for this age group, as it provides greater reassurance through the likelihood of a double negative result While HPV infections are more common among younger sexually active women, the incidence of cervical cancer in this demographic remains low.

HPV testing in women aged 30 and older has high sensitivity but decreased specificity, leading to the detection of many non-carcinogenic HPV cases and unnecessary colposcopy referrals Negative co-testing results in this age group indicate a low risk of developing CIN 2 or CIN 3 within five years, allowing for a safe extension of the screening interval from three to five years While co-testing facilitates earlier detection of abnormal findings prior to a cervical cancer diagnosis, it has faced criticism for potentially increasing the number of tests, referrals to colposcopy, and overall healthcare costs.

The selected records primarily utilized data from clinical trials, which incorporated reminder calls and follow-ups that surpass the current standard of care at the population level, potentially leading to slight cost increases Additionally, the clinical trial data did not account for the management and treatment costs associated with CINs and CC Relevant cost inputs were inferred by adjusting data from other countries with similar cases, although this extrapolation may introduce bias into the analysis.

Selected countries, including the Netherlands, USA, Ireland, Korea, Germany, Greece, British Columbia, Canada, Australia, and China, have implemented HPV vaccination programs for girls and women, with the exception of China In populations where both vaccinated and unvaccinated women are present, the optimal screening strategy remains unclear Vaccinated women, particularly in the youngest cohorts with limited herd immunity, face significantly lower risks compared to their unvaccinated counterparts However, some studies have hesitated to address the influence of herd immunity on the effectiveness of screening tests for unvaccinated women Therefore, it is essential to establish recommendations for screening strategies based on vaccination status to ensure validation and effectiveness.

Many studies have opted for cost-saving measures and differences in colorectal cancer (CC) incidence for comparisons instead of calculating an Incremental Cost-Effectiveness Ratio (ICER) This approach is particularly relevant in countries like Australia, the US, Ireland, Greece, Germany, British Columbia, Canada, and Korea, where national CC screening programs are already in place Consequently, these investigations prioritize screening strategies that maximize cost savings and the reduction in CC-related deaths and incidence In one US study, the authors utilized thresholds of $50,000 to $200,000 per Quality-Adjusted Life Year (QALY) gained, while two studies in China and the US employed no screening scenarios as comparators to evaluate cost-effectiveness However, these comparators may be inappropriate, potentially introducing biases that could affect the outcomes and recommendations of economic analyses.

Recent studies indicate that co-testing is more cost-effective in reducing cervical cancer (CC) incidence and mortality compared to cytology alone However, the difference in outcomes when using HPV testing as a benchmark was not significant Co-testing resulted in the lowest CC incidence rates and demonstrated high efficacy in early detection Although it yielded a greater number of Quality-Adjusted Life Years (QALYs), the cost per screened woman was notably higher, raising concerns for policymakers Consequently, many studies suggest that HPV-based screening strategies are more economically viable due to lower costs It is important to note that these findings stem from populations with high HPV vaccination rates, where cytology may be less effective, potentially explaining the minimal differences in sensitivity and specificity between co-testing and HPV testing Among the 13 studies reviewed, only two from the US and Ireland confirmed the cost-effectiveness of co-testing.

A recent study by Juan C Felix et al in the US, utilizing data from the ATHENA trial with only 2% vaccinated participants, compared the effectiveness of co-testing (Pap smear plus HPV mRNA genotype test) against HPV DNA genotype testing Additionally, an observational cohort study by Binhua Dong et al in China suggested that co-testing with HPV genotype tests may be more cost-effective than combining co-testing with other HPV tests This research advocates for the use of HPV genotype testing as a viable alternative for cervical cancer screening strategies in low- and middle-income countries (LMICs).

In contrast to the research by Binhua Dong et al., findings from Adam Keane et al in China indicated that co-testing was not cost-effective compared to HPV testing, primarily due to the increased number of patients requiring treatment Nevertheless, both methods demonstrated similar effectiveness in preventing cervical cancer (CC) incidence and mortality Notably, the study by Adam Keane et al did not specify the type of HPV test utilized in the cervical cancer strategies involving co-testing, unlike the research conducted by Binhua Dong et al.

Dong et al, implied that different technologies could bring better clinical and economics benefits.

These findings were also consistent with results from literature review in section 1

In unvaccinated communities, relying solely on HPV tests for cervical cancer screening without cytology backup can result in a significant number of false positives Most studies supporting HPV testing as a first-line screening method originate from countries with high HPV vaccination rates since 2000, with China being an exception These studies emphasize the clinical and economic advantages of HPV genotyping in screening strategies for vaccinated populations Recent national guidelines in Europe, the USA, and Australia are shifting from cytology-based tests to HPV testing, with or without genotyping for HPV 16/18 Australia pioneered HPV-based cervical cancer screening in 2017, despite projections indicating a 36% increase in colposcopies among unvaccinated women While high-income countries like Australia and the USA can manage the financial implications of this transition due to a lower percentage of unvaccinated women, low- and middle-income countries face significant financial challenges.

The use of liquid-based cytology (LBC) with high-risk HPV (hr-HPV) testing as a reflex test differs significantly from HPV testing with cytology triage for cervical cancer (CC) screening LBC combined with hr-HPV testing enhances specificity for detecting high-grade lesions, albeit with higher false positive rates Studies have shown that utilizing hr-HPV testing as a reflex triage reduces referral rates to colposcopy and lowers the overall abnormal cytology reporting rate, thereby alleviating unnecessary psychological stress from excessive referrals In contrast, HPV testing with cytology triage tends to have lower sensitivity, as the triage process diminishes the sensitivity benefits of HPV testing, leading to fewer detected cases of CC and cervical intraepithelial neoplasia (CIN) However, this method may exhibit higher relative specificity Notably, research indicates that HPV testing with cytology triage requires fewer colposcopies to identify one CIN2 lesion or more severe conditions compared to the reflex testing approach.

Main findings from the CEA of the co-testing method

The analysis indicates that cervical cancer (CC) screening through three consecutive co-testing sessions is less cost-effective than screening through five cytology tests This finding contradicts existing literature, potentially due to differences in screening frequency among the target population Previous studies modeled CC screening throughout a woman's life until a specified age, while this study specifically examines the effectiveness of three versus five screenings, resulting in older women (over 49) receiving fewer screenings, particularly with the co-testing method Notably, the incidence of CC peaks in Vietnamese women over 45, leading to reduced Quality-Adjusted Life Years (QALY) in older cohorts Additionally, discrepancies in the cytology effectiveness cut-off points may contribute to the differences observed This study used a pooled sensitivity for high-grade squamous intraepithelial lesions (HSIL or CIN3), whereas other literature did not specify their cut-off points, often relying on data from mixed populations in clinical trials, which may further lower cytology sensitivity.

A study on the DSA revealed that the Incremental Cost-Effectiveness Ratio (ICER) is most influenced by the transition probability from HPV hr(-) to HPV hr(+), the prevalence of HPV hr in the general female population, and the frequency of cervical cancer (CC) screenings While the transition probability cannot be altered, the other factors can be improved through HPV vaccination programs and enhanced CC screening strategies The Vietnamese government could facilitate HPV vaccine coverage for girls by negotiating the price to approximately $4.55 per dose The first two factors keep the ICER within the World Health Organization's recommended threshold for low- and middle-income countries (LMIC) However, increasing routine CC screenings could shift the cost-effectiveness analysis from unfavorable to favorable, suggesting that cytology screening may become more cost-effective than co-testing with fewer screenings Future research should focus on determining the optimal number of cytology screenings required to surpass the cost-effectiveness of the co-testing method.

The co-testing method for cervical cancer (CC) screening is less cost-effective compared to the cytology method across all age groups, resulting in higher costs and lower quality-adjusted life years (QALY) gained If the intervention and comparator were reversed, potential savings could range from $16 to $64 per QALY gained However, the incremental cost-effectiveness ratio (ICER) for three instances of co-testing versus three instances of cytology stands at $152 per QALY gained, indicating that women would need to spend $152 to achieve one additional QALY, as opposed to saving money with the cytology method While co-testing may reduce the incidence of CC cases, it comes with significantly higher costs and an increased rate of false positives, leading to additional financial burdens for triage and follow-up testing These findings align with existing literature on the cost-effectiveness of co-testing, which suggests that while it may offer medical benefits, those benefits come at a substantially higher cost.

The study identifies several limitations regarding the proposed age interval for cervical cancer screening in Vietnam, suggesting a range of 25 to 55 years, while other countries recommend screening for women aged 21 to 65 or even up to 79 years Future research should explore the most appropriate age range for cervical screening in Vietnam If the current study finds that screening women aged 25 to 55 is cost-effective, additional research will be necessary to evaluate the effects on younger and older women Furthermore, the proposed screening timeline is primarily based on WHO recommendations and a systematic review of cost-effectiveness in low- and middle-income countries, highlighting a potential weakness due to a lack of comprehensive evidence.

The research does not account for the impact of HPV vaccination on the transition probabilities within the Markov model Since the HPV vaccine has been available in Vietnam since 2009, the prevalence of vaccinated women in the target group remains unknown, potentially affecting the sensitivity and positive predictive value of the cytology method Consequently, the researcher reviewed literature on the efficacy and effectiveness of cervical cytology in low- and middle-income countries (LMIC) to refine the input parameters for the effectiveness of cervical cancer screening methods.

The Markov model in the study utilizes 21 transition probabilities, based on a ± 25% range assumption from a Chinese study on cervical cancer elimination A limitation arises from applying the same transition probabilities across different age groups, although sensitivity analyses confirmed their similarity The researcher validated the transition probabilities with four Vietnamese clinical experts to ensure robust evidence Additionally, the study's QALY weights were derived from a large population in Canada and the USA using the standard gamble method, despite the EQ-5D being the most reliable QALY measurement tool The literature review indicated minimal differences between QALY weights from both methods, supporting the use of Warner's data for this study.

K H et al (2015) was used They still contribute to the limitations of the study and the future studies should investigate the transition probabilities and QALY weights of CC from the Vietnamese context

To address uncertainties in this study, the researcher employed probabilistic sensitivity analysis (PSA) However, the original study on the Markov model's transition probabilities did not provide essential statistical parameters such as standard deviation, 95% confidence intervals, or distribution details The researcher estimated these values based on transition probabilities across age groups and data from the study “A dynamic Bayesian Markov model for health economic evaluations of interventions in infectious disease,” generating a beta distribution for the PSA While applying the same distribution across all age groups presents a limitation, it was deemed the best available practice Future research should focus on systematic reviews and meta-analyses concerning the transition probabilities of colorectal cancer development in the general population.

Conclusion and recommendation

In the past century, significant advancements in cervical cancer screening have led to earlier detection and improved outcomes for women Ongoing research into the disease's biology has generated various screening trials and cost-effectiveness analyses Deciding on the most suitable cervical cancer screening strategies depends on regional resources, implementation feasibility, and acceptable risk levels A thorough comparison of clinical methods and economic evaluations is essential for making informed decisions Although HPV tests offer greater reassurance when negative compared to cytology alone, their lower specificity can result in unnecessary referrals and heightened patient anxiety While co-testing may address the limitations of both methods, it is generally not cost-effective at the population level due to its minimal advantages over other screening options.

Since the introduction of the Pap smear by Dr Papanicolaou, cervical cytology has proven effective in detecting cervical lesions through routine microscopy The Pap test, enhanced over the years with liquid-based cytology (LBC), offers superior specificity but lower sensitivity compared to HPV testing In high-income countries, the increased HPV vaccination coverage among girls and women over the last decade may reduce cytology sensitivity due to a decline in the prevalence of abnormal lesions, leading to a preference for HPV-based screening strategies that provide both clinical and economic advantages, as supported by numerous cost-effectiveness analyses However, the reliability of HPV test results can result in unnecessary referrals for confirmatory tests like colposcopies and biopsies, especially since not all HPV test platforms have FDA approval; thus, studies that do not specify the HPV platform should be approached with caution Incorporating additional cytology tests can enhance the sensitivity of HPV testing and help identify non-HPV-related infections and lesions Utilizing a combination of these tests—either HPV testing with cytology triage or vice versa—across different age groups may yield the most effective cervical cancer screening strategy in regions where both tests are accessible Additionally, limited resources and costs in low- and middle-income countries pose challenges in establishing an appropriate cervical cancer screening strategy.

Three consecutive cervical cancer screenings using the co-testing method are less cost-effective than five screenings using the cytology method While the co-testing approach may be cost-effective in certain scenarios, offering savings of 16 to 64 USD per QALY gained, it significantly increases unnecessary referrals and treatments for healthy women This not only poses a financial burden on Vietnam's healthcare system but also raises mental health concerns for patients Due to the higher costs and lower QALY gains associated with three consecutive screenings via co-testing, this method is not recommended for Vietnamese women.

25 to 55 years old With reasonable clinical benefits and saving cost per QALYs gained, the strategy using 5 times consecutive CC screening by cytology can take into consideration

Annex 1: Consent form for expert consultation 5

I willingly agree to participate in this research study, fully understanding that I can withdraw at any time without facing any negative repercussions.

I have been made aware of the study's purpose and the session details I have read the participant information sheet or had it read to me, and I was given the chance to ask questions, all of which were answered to my satisfaction.

My personal information will remain confidential and anonymous, accessible only to the investigators I have received contact details for individuals to reach out to if I have any questions regarding the study.

_ _ _ D D M M Y Y Printed name of participant Signature of participant

Thumbprint if participant is illiterate

For person who obtained written informed consent:

Printed name of witness Signature of witness Date of signature

5 Adopted from the consent form of the study “Situation Analysis of Migrant Health in Viet Nam” (2020)

Annex 2: Interview guide for consulting with cancer experts

Thank you for participating in this interview and introduce yourself (interviewer)

We are reaching out to seek your expert advice on the natural history and treatment scenarios of cervical cancer Your insights would be invaluable to our study, and we greatly appreciate your time and contribution.

The interview will last approximately one hour and participation is completely voluntary, allowing you to withdraw at any time for any reason Your personal information will be coded and anonymized to ensure your identity remains confidential The findings from this study aim to evaluate the cost-effectiveness of cervical screening methods within the context of Vietnam.

Note Taking and Tape Recording

Today, we will be discussing important topics, and while will take notes, we will also tape-record the conversation for accuracy The recording will be reviewed later to ensure a complete transcript of our discussion.

This confidential tape-recorded conversation aims to accurately document your statements and opinions All personal information will be securely stored, and the research team will not disclose any details to outside parties We kindly request that participants refrain from discussing the content of this meeting with others Please note that all records will be destroyed after September 2022.

Note: The order of questions is only a suggestion Based on the actual situation The interviewer needs to operate flexibly to exploit information

1 General information: Please introduce about yourself and your working experience related to cervical cancer

- Could you please describe the cause and natural history development of cervical cancer?

- In your opinion, what aspects of the Markov model should be revised? Please explain the reason

3 Effectiveness of cervical cancer screening method:

The specificity and sensitivity of cervical cancer screening methods require careful revision based on recent literature findings It is essential to address the discrepancies in reported sensitivity rates, which can vary significantly among different populations and screening techniques A suggested sensitivity range of 70-90% for Pap smears and 90-95% for HPV testing could provide a more accurate reflection of current practices Additionally, the specificity of these methods, often reported between 80-95%, should be standardized to enhance comparative studies By refining these parameters, we can improve the accuracy of cervical cancer screening and ultimately enhance patient outcomes.

4 Treatment scenario for each health stage

- Please list out all medical services related to cervical cancer treatment from HPV infected period to cervical cancer

- Please tell us treatment regimen/scenario for each health stage in the cervical cancer development process (HPV hr(+); CIN 1,2,3; CC)

- In your opinion, what aspects of the treatment regimen/scenario from the literature review should be revised? Please explain the reason

5 Transition probability of health stages

The literature review on the transition probabilities of health stages related to cervical cancer reveals several areas that require revision Firstly, the current probabilities may not accurately reflect the latest advancements in screening and vaccination, which have significantly altered disease progression Additionally, the inclusion of diverse populations in studies is essential to ensure that the probabilities are representative and applicable across different demographics It is suggested that probabilities be updated to account for these factors, incorporating recent data on survival rates and treatment efficacy to provide a more accurate reflection of the disease's progression.

Based on the literature review, it is essential to revise the aspects of patient QALYs (Quality-Adjusted Life Years) related to cervical cancer across different health stages This revision is necessary to ensure that the metrics accurately reflect the varying impacts of treatment and disease progression on patients' quality of life Adjustments should include incorporating patient-reported outcomes and considering the psychological and social dimensions of living with cervical cancer, as these factors significantly influence overall well-being Additionally, refining the measurement tools to capture nuances in health-related quality of life at different disease stages can lead to more precise evaluations and better-informed healthcare decisions.

Annex 3: Interview guide for consulting with reproductive health expert

Thank you for participating in this interview and introduce yourself (interviewer)

We are reaching out for your expert advice on the natural history and treatment scenarios of cervical cancer, as previously discussed in our email regarding the study Your insights would be invaluable, and we greatly appreciate your time and assistance.

The interview, which is voluntary and can be exited at any time, is designed to last no longer than an hour Your personal information will be coded and anonymized to ensure your identity remains confidential The findings from this study will offer valuable insights into the cost-effectiveness of cervical screening methods within the context of Vietnam.

Note Taking and Tape Recording

Tiêu đề	Cervical Cancer Screening by Co-Testing Method for Vietnamese Women from 25 to 55 Years Old: A Cost – Effectiveness Analysis
Tác giả	Bui Thu Hien
Người hướng dẫn	PhD. Pham Nu Hanh Van, Assoc. Prof. Vu Hong Thang
Trường học	Hanoi University of Public Health
Chuyên ngành	Master of Public Health
Thể loại	Thesis
Năm xuất bản	2022
Thành phố	Hanoi

Định dạng
Số trang	126
Dung lượng	20,65 MB