Abstract
Objective
Self-sampled human papillomavirus (HPV) testing is a potential option for cervical cancer screening, but research is scarce in Cambodia. We evaluated the feasibility, accuracy, and acceptability of self-sampled HPV testing using careHPV.
Methods
A cross-sectional study including women aged 20–49 years attending 2 national hospitals in the capital city was conducted. Women underwent both self-sampling and clinician-sampling of specimens, and were then asked to complete an acceptability questionnaire. The paired samples were analyzed for high-risk HPV by careHPV and genotyped by polymerase chain reaction (PCR).
Results
A total of 375 women were eligible for inclusion. Based on PCR, 78.9% were negative for HPV in both self and clinician-samples, 9.9% had a complete HPV type match, and 6.1% had all HPV types in clinician-samples also detected in self-samples. In 5.1%, one or more HPV types identified in the clinician-samples were missed in self-samples. When using careHPV, the overall agreement between the 2 sampling methods was 95.7% (95% confidence interval [CI]=95.8–95.6) with good concordance (κ=0.66, 95% CI=0.56–0.76). Nearly 90% of the women preferred clinician-sampling over self-sampling, citing greater comfort, ease, and speed.
Conclusion
Self-sampled HPV testing using careHPV could be an option for cervical cancer screening in Cambodia; however, it requires periodic quality control of handling procedures. In addition, women’s health education regarding the accuracy of self-sampled HPV testing and the importance of follow-up in cases of positive results is needed.
Keywords: Cambodia, Uterine Cervical Neoplasms, Early Detection of Cancer Papillomavirus Infections
Synopsis
We examined the feasibility, accuracy, and acceptability of self-sampled human papillomavirus (HPV) testing in Cambodia. The majority of women were able to perform self-sampling and correctly collect HPV-infected samples. A good concordance was seen between the careHPV results of self- and clinician-samples. Women preferred clinician-sampling.
INTRODUCTION
Despite its preventability, cervical cancer continues to be one of the leading cancers affecting women in Cambodia. The estimated age-standardized incidence rate of cervical cancer is 14.0 per 100,000 and the mortality rate is 8.3 per 100,000, which are both higher than the global estimates [1]. The National Action Plan for Cervical Cancer Prevention and Control 2019–2023 includes introducing the human papillomavirus (HPV) vaccine into the routine immunization schedule, but the vaccine has not been introduced to date (n.b. it is planned for introduction from October 2023) [2]. Cervical cancer screening using visual inspection with acetic acid (VIA) is available at some health centers and Pap smear at national hospitals in the capital city of Phnom Penh; however, they are only offered opportunistically. Scaling up screening services is crucial, but concerns have been raised about the accuracy of VIA, which relies solely on subjective interpretation [3]. Pap smear testing is challenging to expand nationwide due to the shortage of pathology workforce [4]. In light of this situation, we conducted a screening project using HPV testing using careHPV (Qiagen, Hilden, Germany) between 2015 and 2018 [3,5]. This project confirmed that the testing could be effectively performed in a Cambodian setting. However, the low participation rate of only 19% remains a significant challenge.
Growing evidence suggests that self-sampled HPV testing, in which a woman collects cervical-vaginal samples by herself, can be an alternative approach to clinician-sampled testing. In 2 systematic reviews, self-sampling increased screening uptake when available in addition to clinician sampling as a choice [6,7]. Accuracy is a concern with self-sampled HPV testing using careHPV, as it has been shown to have lower sensitivity for detecting cervical intraepithelial neoplasia (CIN) 2 or worse compared to a polymerase chain reaction (PCR) [6]. However, studies from low- and middle-income countries have reported that self-sampled HPV testing using careHPV has higher sensitivity than VIA or cytology, making it an effective screening method [8,9]. This is especially relevant in settings where PCR is not readily available or affordable.
In Cambodia, there is one study examining self-sampled HPV testing using careHPV [10]. However, it remains unclear whether women can perform self-sampling correctly, whether careHPV can be used, and whether they favorably accept this new method. This study evaluated the feasibility, accuracy, and acceptability of self-sampled HPV testing using careHPV.
MATERIALS AND METHODS
1. Study design
This was a cross-sectional study at 2 national hospitals, the National Maternal Child Health Center and Khmer-Soviet Friendship Hospital, located in Phnom Penh, the capital city of Cambodia. These 2 hospitals are among the few that offer HPV testing in Cambodia.
2. Study participants
Women aged 20–49 years who visited the gynecology outpatient department of 2 hospitals between June 2021 and June 2022 were recruited for this study. Those who were pregnant, had a history of total hysterectomy, or required urgent medical care were excluded. We aimed to recruit as many women as possible with HPV infection; we therefore included women in their 20s, while acknowledging that the HPV-positive status of such patients tends to be largely clinically insignificant. We did not evaluate the test in women with certain conditions, such as cytologic or histologic abnormalities, immunocompromised status, or a history of sexually transmitted diseases.
3. Sample size calculation
In our previous study, the positivity rate of clinician-sampled HPV testing was found to be 5% among Cambodian women [5]. To determine the target sample size for this study, we examined how the Kappa coefficient would change if the HPV positivity rate was 4% or 7%, agreement rate was 90% or 95%, or sample size was 300, 500, or 1,000. The Kappa coefficients were similar regardless of sample size once HPV positivity and agreements rates were accounted for. For a feasible study design, the target sample size was set at 300. Because both hospitals were designated facilities for coronavirus disease 2019 patients, routine outpatient care, including cervical cancer screening, was suspended during the infection outbreak; thus, it took longer than initially planned to obtain data on the target sample size.
4. Sample collection
Women who agreed to participate in the study were interviewed regarding their sociodemographic characteristics, including marital status, reproductive history, and educational background. They were provided with visual and oral instructions on how to correctly perform self-sampling of cervico-vaginal specimens and then given privacy behind a partition in an examination room to collect their specimens using the careBrush and careHPV collection media. Additionally, a specimen was collected by a gynecologist for comparison. Following the samplings, participants were asked to complete an acceptability questionnaire.
5. HPV testing
At the laboratory of the 2 hospitals, the collected samples were stored at 4°C. Within one week of collection, a trained laboratory technician analyzed samples for high-risk HPV (hrHPV) using the careHPV test (Qiagen). The careHPV test is a signal amplification assay that utilizes microplate chemiluminescence for the qualitative detection of 14 hrHPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68) [11,12]. The system is based on Hybrid Capture 2 technology and is prequalified by the World Health Organization (WHO) for primary use in low-resource settings, due to its ability to operate without running water or stable electricity. A cutoff of 1 pg/mL was used to define test positivity.
HPV testing was also conducted by PCR using the PGMY-CHUV assay at a WHO-certified HPV reference laboratory in Japan [13]. Briefly, the assay is performed using the PGMY09/11 L1 consensus primer set, followed by reverse blotting hybridization with oligonucleotide probes specific for 31 HPV genotypes (HPV-6, 11, 16, 18, 26, 31, 33, 34, 35, 39, 40, 42, 44, 45, 51, 52, 53, 54, 55, 56, 57, 58, 59, 66, 68, 69, 70, 73, 82, 83 and 84) [14,15].
6. Questionnaire
The authors developed a questionnaire in English based on previous studies [16,17]. This English original was translated into Khmer, then back-translated into English to check for consistency, and the final version of the questionnaire was pre-tested on some women. It included questions on the ease of understanding self-sampling instructions, confidence in performing self-sampling, pain, embarrassment, concerns about the accuracy of self-sampling and clinician-sampling methods, and method of preference. The reasons for preferring a selected method were also asked.
7. Data analysis
Two research staff members collected all data on paper and entered it into a Microsoft Excel sheet. Data analysis was performed using Stata SE16 software (Stata Corporation, College Station, TX, USA). Statistical significance was set at p≤0.05.
We analyzed the sociodemographic characteristics of the study participants using descriptive statistics. To assess the feasibility of self-sampling, we compared the PCR test results of self-collected samples to those collected by a clinician, evaluating whether HPV types infecting the cervix could be correctly detected in self-samples. The accuracy of self-sampled HPV testing using careHPV was examined by the concordance in careHPV test results between paired self- and clinician-samples. The concordance was assessed by calculating the overall agreement rate and Cohen’s kappa statistics with the following interpretation: poor or slight agreement (≤0.2), fair agreement (0.2–0.4), moderate agreement (0.4–0.6), good agreement (0.6–0.8), and very good agreement (0.8–1.0) [18]. In addition, the sensitivity and specificity of careHPV for hrHPV detection was examined using PCR as the gold standard. In PCR, hrHPV positivity was defined as when at least one of the 14 HPV types was detected (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68). Paired t-tests were used to compare pain, embarrassment, and concerns regarding the accuracy of the results between the 2 sampling methods. Pearson’s χ2 test was used to examine sampling preferences based on sociodemographic characteristics.
8. Ethical considerations
The study protocol was approved by the Cambodia National Ethics Committee for Health (295 NECHR) and the ethical committees of the National Center for Global Health and Medicine and Nihon University School of Medicine, Japan (NCGM-G-003613-00 and RK-201013-4, respectively). Written informed consent was obtained from all the participants.
RESULTS
1. Patient characteristics
We recruited a total of 376 women, of which 375 (99.7%) were eligible for analysis. One woman in her 30s was excluded from the analysis because she experienced fear during self-sampling and could not perform the procedure. Table 1 presents the sociodemographic characteristics of the study participants. The mean age was 37.5±6.8 years. The majority of women were married (95.2%), had 2 or more children (81.6%), and had graduated from high school or higher education (70.1%). More than half of the participants were employees.
Table 1. Sociodemographic characteristics of the study participants.
| Characteristics | No. (%) | |
|---|---|---|
| Total | 375 | |
| Age (yr) | ||
| Mean ± standard deviation | 37.5±6.8 | |
| 20–29 | 47 (12.5) | |
| 30–39 | 186 (49.6) | |
| 40–49 | 142 (37.9) | |
| Marital status | ||
| Married | 357 (95.2) | |
| Divorced | 13 (3.5) | |
| Widowed | 5 (1.3) | |
| Parity | ||
| 0 | 16 (4.3) | |
| 1 | 53 (14.1) | |
| ≥2 | 306 (81.6) | |
| Menstrual bleeding pattern | ||
| Regular | 215 (57.3) | |
| Irregular | 144 (38.4) | |
| Menopause | 16 (4.3) | |
| Highest level of school attended | ||
| Primary school | 33 (8.8) | |
| Secondary school | 73 (19.5) | |
| High school | 165 (44.0) | |
| College or higher | 98 (26.1) | |
| Other | 6 (1.6) | |
| Employment status | ||
| Employee | 218 (58.1) | |
| Unemployed | 81 (21.6) | |
| Other | 76 (20.3) | |
| Tobacco use | ||
| Never | 375 (100.0) | |
| Current or ex-smoker | 0 (0.0) | |
2. Feasibility of self-sampling
Table 2 summarizes the PCR test results obtained using the 2 sampling methods. Among 375 women, 296 (78.9%) were negative for HPV in both self and clinician-samples. Of the remaining women with HPV detected in either sample, 37 (9.9%) had a complete HPV type match, and 23 (6.1%) had all HPV types in their clinician-samples (i.e., infecting the cervix) also detected in self-samples. In 19 women (5.1%), one or more HPV types identified in the clinician-samples were missed in self-samples, including 4 women with HPV-16. No age difference was observed between women in whom all HPV infections in the cervix were detected and those in whom they were not (p=0.10).
Table 2. HPV types detected in self- and clinician-samples.
| Characteristics | Self-samples | Clinician-samples | Subtotal No. | No. (%) |
|---|---|---|---|---|
| Negative in both samples | Negative | Negative | - | 296 (78.9) |
| Complete match of HPV types in both samples | HPV-6, 11 | HPV-6, 11 | 1 | 37 (9.9) |
| HPV-11 | HPV-11 | 1 | ||
| HPV-16 | HPV-16 | 5 | ||
| HPV-18 | HPV-18 | 3 | ||
| HPV-18, 44 | HPV-18, 44 | 1 | ||
| HPV-33 | HPV-33 | 3 | ||
| HPV-42 | HPV-42 | 1 | ||
| HPV-44 | HPV-44 | 3 | ||
| HPV-45 | HPV-45 | 1 | ||
| HPV-51 | HPV-51 | 2 | ||
| HPV-52 | HPV-52 | 3 | ||
| HPV-53 | HPV-53 | 1 | ||
| HPV-53, 55 | HPV-53, 55 | 1 | ||
| HPV-55, 56 | HPV-55, 56 | 1 | ||
| HPV-56 | HPV-56 | 2 | ||
| HPV-58 | HPV-58 | 3 | ||
| HPV-59 | HPV-59 | 2 | ||
| HPV-68 | HPV-68 | 2 | ||
| HPV-70 | HPV-70 | 1 | ||
| All HPV types detected in clinician-samples also detected in the self-samples | HPV-16, 18, 40, 42, 66 | HPV-16, 18, 40 | 1 | 23 (6.1) |
| HPV-16, 18, 45, 58, 66 | HPV-16, 18, 45, 66 | 1 | ||
| HPV-16, 68 | HPV-16 | 1 | ||
| HPV-42, 44, 52, 55 | HPV-55 | 1 | ||
| HPV-44, 68 | HPV-68 | 1 | ||
| HPV-53, 59 | HPV-53 | 1 | ||
| HPV-58, 59 | HPV-59 | 1 | ||
| HPV-16 | Negative | 6 | ||
| HPV-44 | Negative | 3 | ||
| HPV-51 | Negative | 2 | ||
| HPV-52 | Negative | 2 | ||
| HPV-58 | Negative | 1 | ||
| HPV-68 | Negative | 2 | ||
| One or more HPV types detected in clinician-samples missed in self-samples | HPV-16, 26, 66 | HPV-26, 66, 68 | 1 | 19 (5.1) |
| HPV-16, 18, 44, 52, 56 | HPV-16, 18, 44, 52, 55 | 1 | ||
| HPV-18 | HPV-51 | 1 | ||
| HPV-18 | HPV-68 | 1 | ||
| HPV-31, 66 | HPV-56, 66 | 1 | ||
| HPV-39, 52 | HPV-39, 53 | 1 | ||
| HPV-52, 53 | HPV-53, 54 | 1 | ||
| HPV-53 | HPV-39, 53 | 1 | ||
| HPV-59 | HPV-52, 59 | 1 | ||
| HPV-44 | HPV-16 | 1 | ||
| Negative | HPV-16 | 3 | ||
| Negative | HPV-40 | 1 | ||
| Negative | HPV-45 | 1 | ||
| Negative | HPV-58 | 1 | ||
| Negative | HPV-68 | 1 | ||
| Negative | HPV-70 | 2 |
HPV, human papillomavirus.
3. Accuracy of self-sampled HPV testing using careHPV
Table 3 presents a contingency table comparing careHPV test results between the self- and clinician samples. When using careHPV, both the self- and clinician-samples tested positive for hrHPV in 17 women (4.5%) and negative in 342 women (91.2%), giving an overall agreement of 95.7% (95% confidence interval [CI]=95.8–95.6) and good concordance (κ=0.66, 95% CI=0.56–0.76) (Table 3). The positive and negative percent agreement were 65.4% and 98.0%, respectively.
Table 3. Contingency table comparing careHPV test results between self- and clinician-samples.
| Clinician-samples | Self-samples | ||
|---|---|---|---|
| hrHPV positive | hrHPV negative | Total | |
| hrHPV positive | 17 (4.5) | 9 (2.4) | 26 (6.9) |
| hrHPV negative | 7 (1.9) | 342 (91.2) | 349 (93.1) |
| Total | 24 (6.4) | 351 (93.6) | 375 (100.0) |
hrHPV, high-risk human papillomavirus.
Table 4 shows a contingency table comparing careHPV and PCR results for hrHPV detection using the 2 sampling methods. In self-samples, the sensitivity and specificity of careHPV for hrHPV detection were 41.1% (95% CI=28.1–55.0) and 99.7% (98.3%–100.0%), respectively. In clinician samples, the sensitivity was only slightly higher.
Table 4. Contingency table of careHPV and PCR results for hrHPV detection by sampling method.
| Characteristics | careHPV | ||||
|---|---|---|---|---|---|
| Self-samples | Clinician-samples | ||||
| hrHPV positive | hrHPV negative | hrHPV positive | hrHPV negative | ||
| PCR using PGMY-CHUV assay | |||||
| hrHPV positive | 23 | 33 | 22 | 26 | |
| hrHPV negative | 1 | 318 | 4 | 323 | |
| Total | 24 | 351 | 26 | 349 | |
| Sensitivity % (95% CI) | 41.1 (28.1–55.0) | 45.8 (31.4–60.8) | |||
| Specificity % (95% CI) | 99.7 (98.3–100.0) | 98.8 (96.9–99.7) | |||
CI, confidence interval, hrHPV, high-risk human papillomavirus, PCR, polymerase chain reaction.
4. Acceptability of self-sampling
Table 5 presents the results of the acceptance questionnaires. Among 375 participants, 90.4% reported that the instructions for self-sampling were easy or very easy to understand, and 80.3% felt confident about collecting the sample by themselves. There was no difference in pain between the sampling methods. However, women felt significantly less embarrassment (9.4% vs. 21.3%, p<0.001) and were more concerned about the accuracy of the test results (42.7% vs. 17.3%, p<0.001) with self-sampling. Regarding participant preferences, 89.9% preferred clinician-sampling over self-sampling, citing greater comfort, ease, and speed. No significant differences in preferences were observed according to sociodemographic characteristics (Table 6).
Table 5. Acceptability of self-sampling versus clinician-sampling.
| Characteristics | Self-sampling | Clinician-sampling | p-value (paired t-test) | |
|---|---|---|---|---|
| Understanding of self-sampling instructions | NA | NA | ||
| Very easy | 58 (15.5) | |||
| Easy | 281 (74.9) | |||
| Difficult | 25 (6.7) | |||
| Very difficult | 11 (2.9) | |||
| Confidence in self-sampling | NA | NA | ||
| Yes | 301 (80.3) | |||
| No | 74 (19.7) | |||
| Pain | 0.880 | |||
| No pain at all | 125 (33.3) | 116 (30.9) | ||
| Not so painful | 225 (60.0) | 243 (64.8) | ||
| Painful | 21 (5.6) | 14 (3.7) | ||
| Very painful | 3 (0.8) | 1 (0.3) | ||
| No answer | 1 (0.3) | 1 (0.3) | ||
| Embarrassment | <0.001 | |||
| Not embarrassing at all | 136 (36.3) | 75 (20.0) | ||
| Not so embarrassing | 204 (54.4) | 219 (58.4) | ||
| Embarrassing | 34 (9.1) | 74 (19.7) | ||
| Very embarrassing | 1 (0.3) | 6 (1.6) | ||
| No answer | 0 (0) | 1 (0.3) | ||
| Concern about accuracy of the result | <0.001 | |||
| No concern at all | 61 (16.3) | 122 (32.5) | ||
| Not so concerned | 154 (41.1) | 186 (49.6) | ||
| Concerned | 136 (36.3) | 57 (15.2) | ||
| Very concerned | 24 (6.4) | 8 (2.1) | ||
| No answer | 0 (0) | 2 (0.5) | ||
NA, not applicable.
Table 6. Sampling method of preference by sociodemographic characteristics.
| Characteristics | Prefer self-sampling | Prefer clinician-sampling | p-value (χ2 test) | |
|---|---|---|---|---|
| Total | 38 (10.1) | 337 (89.9) | NA | |
| Age (yr) | 0.6 | |||
| 20–29 | 5 (10.6) | 42 (89.4) | ||
| 30–39 | 16 (8.6) | 170 (91.4) | ||
| 40–49 | 17 (12.0) | 125 (88.0) | ||
| No. of delivery | 0.06 | |||
| 0 | 3 (18.8) | 13 (81.3) | ||
| 1 | 1 (1.9) | 52 (98.1) | ||
| ≥2 | 34 (11.1) | 272 (88.9) | ||
| Highest level of school attended | 0.6 | |||
| Primary school | 3 (9.1) | 30 (90.9) | ||
| Secondary school | 11 (15.1) | 62 (84.9) | ||
| High school | 15 (9.1) | 150 (90.9) | ||
| College or higher | 9 (9.2) | 89 (90.8) | ||
| Others | 0 (0.0) | 6 (100.0) | ||
| Employment status | 0.7 | |||
| Employee (healthcare worker) | 6 (7.5) | 74 (92.5) | ||
| Employee (other) | 17 (12.3) | 121 (87.7) | ||
| Unemployed | 7 (8.6) | 74 (91.4) | ||
| Others | 8 (10.5) | 68 (89.5) | ||
NA, not applicable.
DISCUSSION
This study examined whether self-sampled HPV testing using careHPV is feasible, accurate, and acceptable for cervical cancer screening in Cambodia. Our findings suggest that self-sampled HPV testing using careHPV is a valid option, although the sensitivity of careHPV for hrHPV detection is low compared to PCR. In addition, the women in this study preferred clinician-sampling to self-sampling.
The advantage of HPV testing is that it detects viral nucleic acid shed from the cervical and vaginal mucosa into the vaginal cavity, rather than morphological changes in the cell, and thus does not require sampling from the transformation zone of the cervix [19]. In our study, we observed that most women could perform self-sampling under supervision and correctly collect HPV-infected cervical samples, which is encouraging considering that Cambodian women often have limited knowledge of their reproductive anatomy. Although some studies suggest that a woman’s age may affect the quality of HPV self-sampling, we did not find any such association [20].
Concordance or agreement statistics of HPV test results between self- and clinician-samples have been proposed for test validation of self-sampling [21]. Arbyn et al. [21], in their meta-analysis of 7 studies using Hybrid Capture II, reported a pooled overall agreement of 86.7% (95% CI=82.3–90.6), kappa statistics of 0.65 (95% CI=0.58–0.73), positive agreement of 77.7% (95% CI=67.7–86.4), and negative agreement of 91.3% (95% CI=86.9–94.8). One study using careHPV in Tanzania reported an overall agreement of 90.5% (95% CI=87.5–93.0) and kappa statistics of 0.66 (95% CI=0.56–0.75) [20]. The concordance in careHPV test results between self and clinician samples in our study was also good. However, we found that the sensitivity of careHPV for hrHPV detection was less than 50% compared with PCR, regardless of the sampling method. This is not surprising because the PGMY-CHUV assay is a more sensitive method that can detect 31 types of HPV than careHPV, which can detect only 13 types. Further, the PGMY-CHUV assay was established for HPV surveillance rather than cancer screening [13,22]. On the other hand, the sensitivity of careHPV in clinician-samples was slightly lower in this study than in our previous study (53.3%) conducted in the same laboratory, indicating the need for periodic quality control of handling procedures when careHPV is used [5].
As in previous studies, most women felt that the instructions were easy to understand and could perform self-sampling with confidence, minor pain, and less embarrassment than clinician sampling [16,23]. Approximately 40% of women expressed concern about the accuracy of the results with self-sampling, also comparable with previous studies [16,17,20,24]. A unique finding of our study was that almost 90% of the participants preferred clinician-sampling overall, citing reasons such as comfort, ease, and speed. This preference could be because we recruited women who had already visited the gynecology outpatient department of the target hospital and were expecting a pelvic examination. Since self-sampling is a new method that requires instructions and time to perform correctly, women may have felt overwhelmed by the amount of information and time needed for the self-sampling procedure. Understanding the anatomy of reproductive organs and the accuracy of self-sampled HPV testing is essential for self-sampling [17].
The strength of this study was that we carried out HPV testing not only using careHPV but also PCR on all women. However, this study had several notable limitations. First, it was conducted in a hospital setting with close attention from gynecologists and nurses; therefore, it does not represent the entire or underserved population. As self-sampling can be performed without any physical intervention by a clinician, further studies are required to evaluate its feasibility in an unsupervised setting. Second, the lack of colposcopy or biopsy diagnoses for the participants limited our ability to evaluate the accuracy of careHPV compared to the standard histological outcome. Additionally, although the diagnostic evaluation and treatment of screen-positives is a critical component of screening, one third of the women whose clinician-samples were positive by careHPV (10 out of 26 women) were lost to follow-up despite our attempts to recall them. Among 16 women who returned for colposcopy, 7 showed abnormal findings including 2 with CIN 3; however, these women chose to be treated overseas. To scale up cervical cancer screening in Cambodia, the leadership of the government in guiding and managing not only the testing method but also the entire screening process will be essential [25]. Conducting a cost analysis may help make the most appropriate policy recommendations.
In conclusion, self-sampled HPV testing using careHPV could be an option for primary cervical cancer screening in Cambodia; however, the low sensitivity of careHPV for hrHPV detection must be recognized and requires periodic quality control. In addition, health education for women is needed to disseminate knowledge about the accuracy of self-sampled HPV testing and the importance of follow-up in cases of positive results.
ACKNOWLEDGEMENTS
The authors are sincerely grateful to Prof. Sothy Pech, Dr. Linka Koun, Dr. Rayonnette Krouch, Dr. Maryan Chhit, Dr. Lumpiny Kim, Ms. Angkeavatey Som, and Ms. Naren Kanal for their contributions in data collection. We also appreciate Prof. Yutaka Osuga and Dr. Yukari Uemura for their technical inputs.
Footnotes
Funding: This work was supported by the National Center for Global Health and Medicine Intramural Research Fund (20A08) and the Health Labour Sciences Research Grant (23EA3201).
Conflict of Interest: Tadashi Kimura declares the receipt of grants from Chyugai Pharmaceutics Co., Ltd., Asuka Seiyaku Co., Ltd., and Zeria Shinyaku Pharmaceutics Co. Ltd. He also declares the receipt of honoraria from Bayer pharmaceutics Co., Ltd., Asuka Seiyaku Co., Ltd. and Chyugai Seiyaku Co., Ltd. Other authors have no conflict of interest to disclose.
- Conceptualization: H.S., H.R., F.N., K.K.2
- Data curation: H.S., H.R., K.K.2
- Formal analysis: H.R., K.K.2
- Funding acquisition: H.R., F.N., K.K.2
- Investigation: H.S., H.R., K.K.2
- Methodology: H.S., H.R., K.K.,1 K.K.2
- Project administration: H.S., H.R., U.K., K.K.,1 K.K.2
- Supervision: U.K., F.N., K.T., K.K.,1 K.K.2
- Validation: U.K., K.T., K.K.,1 K.K.2
- Writing - original draft: H.S., H.R.
- Writing - review & editing: H.R., F.N., K.T., K.K.,1 K.K.2
K.K.,1 Kanal Koum; K.K.,2 Kei Kawana.
References
- 1.Ferlay JEM, Lam F, Colombet M, Mery L, Piñeros M, Znaor A, et al. Global cancer observatory: cancer today [Internet] Lyon: International Agency for Research on Cancer; 2020. [cited 2023 Mar 29]. Available from: https://gco.iarc.fr/today. [Google Scholar]
- 2.Ministry of Health (CI) National action plan for cervical cancer prevention and control 2019–2023. Phnom Penh: Ministry of Health; 2019. [Google Scholar]
- 3.Kanal K, Fujita N, Soeung SC, Sim KL, Matsumoto Y, Haruyama R, et al. The cooperation between professional societies contributes to the capacity building and system development for prevention and control of cancer in low- and middle-income countries: the practice of Cervical Cancer Prevention and Control Project in Cambodia. Glob Health Med. 2020;2:48–52. doi: 10.35772/ghm.2020.01000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Akaba H, Fujita N, Stauch G, Matsumoto Y, Wakasa T, Kawahara K, et al. How can we strengthen pathology services in Cambodia? Glob Health Med. 2019;1:110–113. doi: 10.35772/ghm.2019.01023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ueda Y, Kawana K, Yanaihara N, Banno K, Chhit M, Uy K, et al. Development and evaluation of a cervical cancer screening system in Cambodia: a collaborative project of the Cambodian Society of Gynecology and Obstetrics and Japan Society of Obstetrics and Gynecology. J Obstet Gynaecol Res. 2019;45:1260–1267. doi: 10.1111/jog.13968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Arbyn M, Smith SB, Temin S, Sultana F, Castle P, et al. Detecting cervical precancer and reaching underscreened women by using HPV testing on self samples: updated meta-analyses. BMJ. 2018;363:k4823. doi: 10.1136/bmj.k4823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yeh PT, Kennedy CE, de Vuyst H, Narasimhan M. Self-sampling for human papillomavirus (HPV) testing: a systematic review and meta-analysis. BMJ Glob Health. 2019;4:e001351. doi: 10.1136/bmjgh-2018-001351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jeronimo J, Bansil P, Lim J, Peck R, Paul P, Amador JJ, et al. A multicountry evaluation of careHPV testing, visual inspection with acetic acid, and papanicolaou testing for the detection of cervical cancer. Int J Gynecol Cancer. 2014;24:576–585. doi: 10.1097/IGC.0000000000000084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Qiao YL, Sellors JW, Eder PS, Bao YP, Lim JM, Zhao FH, et al. A new HPV-DNA test for cervical-cancer screening in developing regions: a cross-sectional study of clinical accuracy in rural China. Lancet Oncol. 2008;9:929–936. doi: 10.1016/S1470-2045(08)70210-9. [DOI] [PubMed] [Google Scholar]
- 10.Thay S, Goldstein A, Goldstein LS, Govind V, Lim K, Seang C. Prospective cohort study examining cervical cancer screening methods in HIV-positive and HIV-negative Cambodian women: a comparison of human papilloma virus testing, visualization with acetic acid and digital colposcopy. BMJ Open. 2019;9:e026887. doi: 10.1136/bmjopen-2018-026887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.World Health Organization. Public report for careHPV test (PQDx 0085-028-00) [Internet] Geneva: World Health Organization; 2018. [cited 2023 Mar 29]. Available from: https://extranet.who.int/pqweb/sites/default/files/180713_pqpr_pqdx_0085_028_00_carehpv_with_labelling.pdf. [Google Scholar]
- 12.Kelly H, Mayaud P, Segondy M, Pant Pai N, Peeling RW. A systematic review and meta-analysis of studies evaluating the performance of point-of-care tests for human papillomavirus screening. Sex Transm Infect. 2017;93:S36–S45. doi: 10.1136/sextrans-2016-053070. [DOI] [PubMed] [Google Scholar]
- 13.Eklund C, Zhou T, Dillner J WHO Human Papillomavirus Laboratory Network. Global proficiency study of human papillomavirus genotyping. J Clin Microbiol. 2010;48:4147–4155. doi: 10.1128/JCM.00918-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bhatla N, Dar L, Patro AR, Kumar P, Kriplani A, Gulati A, et al. Can human papillomavirus DNA testing of self-collected vaginal samples compare with physician-collected cervical samples and cytology for cervical cancer screening in developing countries? Cancer Epidemiol. 2009;33:446–450. doi: 10.1016/j.canep.2009.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Estrade C, Menoud PA, Nardelli-Haefliger D, Sahli R. Validation of a low-cost human papillomavirus genotyping assay based on PGMY PCR and reverse blotting hybridization with reusable membranes. J Clin Microbiol. 2011;49:3474–3481. doi: 10.1128/JCM.05039-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Aiko KY, Yoko M, Saito OM, Ryoko A, Yasuyo M, Mikiko AS, et al. Accuracy of self-collected human papillomavirus samples from Japanese women with abnormal cervical cytology. J Obstet Gynaecol Res. 2017;43:710–717. doi: 10.1111/jog.13258. [DOI] [PubMed] [Google Scholar]
- 17.Hanley SJ, Fujita H, Yokoyama S, Kunisawa S, Tamakoshi A, Dong P, et al. HPV self-sampling in Japanese women: a feasibility study in a population with limited experience of tampon use. J Med Screen. 2016;23:164–170. doi: 10.1177/0969141315625702. [DOI] [PubMed] [Google Scholar]
- 18.Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20:37–46. [Google Scholar]
- 19.Hawkes D, Keung MH, Huang Y, McDermott TL, Romano J, Saville M, et al. Self-collection for cervical screening programs: from research to reality. Cancers (Basel) 2020;12:1053. doi: 10.3390/cancers12041053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Katanga JJ, Rasch V, Manongi R, Pembe AB, Mwaiselage JD, Kjaer SK. Concordance in HPV detection between self-collected and health provider-collected cervicovaginal samples using careHPV in Tanzanian women. JCO Glob Oncol. 2021;7:985–991. doi: 10.1200/GO.20.00598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Arbyn M, Castle PE, Schiffman M, Wentzensen N, Heckman-Stoddard B, Sahasrabuddhe VV. Meta-analysis of agreement/concordance statistics in studies comparing self- vs clinician-collected samples for HPV testing in cervical cancer screening. Int J Cancer. 2022;151:308–312. doi: 10.1002/ijc.33967. [DOI] [PubMed] [Google Scholar]
- 22.World Health Organization. Human papillomavirus laboratory manual, first edition [Internet] Geneva: World Health Organization; 2010. [cited 2023 Mar 29]. Available from: https://www.who.int/publications/i/item/WHO-IVB-10.12. [Google Scholar]
- 23.Madhivanan P, Nishimura H, Ravi K, Pope B, Coudray M, Arun A, et al. Acceptability and concordance of self- versus clinician-sampling for HPV testing among rural south Indian women. Asian Pac J Cancer Prev. 2021;22:971–976. doi: 10.31557/APJCP.2021.22.3.971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Oranratanaphan S, Termrungruanglert W, Khemapech N. Acceptability of self-sampling HPV testing among Thai women for cervical cancer screening. Asian Pac J Cancer Prev. 2014;15:7437–7441. doi: 10.7314/apjcp.2014.15.17.7437. [DOI] [PubMed] [Google Scholar]
- 25.World Health Organization. WHO report on cancer: setting priorities, investing wisely and providing care for all [Internet] Geneva: World Health Organization; 2020. [cited 2023 Mar 29]. Available from: https://apps.who.int/iris/handle/10665/330745. [Google Scholar]