Addition of digital VIA/VILI to conventional naked-eye examination for triage of HPV-positive women: A study conducted in a low-resource setting.

BACKGROUND: World Health Organization guidelines for cervical cancer screening recommend HPV testing followed by visual inspection with acetic acid (VIA) for triage if HPV positive. In order to improve visual assessment and identification of cervical intraepithelial neoplasia grade 2 and worse (CIN2+), providers may use visual aids such as digital cameras.
OBJECTIVES: To determine whether combined examination by naked-eye and digital VIA (D-VIA) and VILI (D-VILI) improves detection of CIN2+ as compared to the conventional evaluation.
MATERIALS AND METHODS: Women (30-49 years) living in Dschang (West Cameroon) were prospectively invited to a cervical cancer screening campaign. Primary HPV-based screening was followed by VIA/VILI and D-VIA/VILI if HPV-positive. Health care providers independently defined diagnosis (pathological or non-pathological) based on naked-eye VIA/VILI and D-VIA/VILI. Decision to treat was based on combined examination (VIA/VILI and D-VIA/VILI). Cervical biopsy and endocervical curettage were performed in all HPV-positive participants and considered as reference standard. Diagnostic performance of individual and combined naked-eye VIA/VILI and D-VIA/VILI was evaluated. A sample size of 1,500 women was calculated assuming a prevalence of 20% HPV positivity and 10% CIN2+ in HPV-positive women.
RESULTS: Due to the COVID-19 pandemic, the study had to terminate prematurely. A total of 1,081 women with a median age of 40 (IQR 35.5-45) were recruited. HPV positivity was 17.4% (n = 188) and 26 (14.4%) had CIN2+. Naked-eye VIA and D-VIA sensitivities were 80.8% (95% CI 60.6-93.4) and 92.0% (95% CI 74.0-99.0), and specificities were 31.2% (95% CI 24-39.1) and 31.6% (95% CI 24.4-39.6), respectively. The combination of both methods yielded a sensitivity of 92.3% (95% CI 74.9-99.1) and specificity of 23.2% (95% CI 16.8-30.7). A trend towards improved sensitivity was observed, but did not reach statistical significance.
CONCLUSION: Addition of D-VIA/VILI to conventional naked-eye examination may be associated with improved CIN2+ identification. Further studies including a larger sample size are needed to confirm these results.

Introduction

Low-income countries like Cameroon have a high prevalence rate of cervical cancer, representing the second leading cause of death by cancer in women [1-3]. In Cameroon, 33.7 per 100’000 women are affected annually, despite cervical cancer being a highly preventable disease [4].

Cervical cancer screening allows to prevent and detect the disease at a precancerous stage and treat it before it progresses to cancer. However, in low-resource settings, the lack of organized screening due, in part, to the absence of dedicated medical structures and trained staff contributes to the problem. Advances in technology for cervical detection in low-resource settings using human papillomavirus (HPV) primary testing followed by immediate treatment with cryotherapy or thermal ablation if needed are adapted to low-resource contexts [5-7] as a ‘screen-and-treat’ approach in a single visit requires fewer resources and reduces risk of loss to follow-up.

Recently, the World Health Organization (WHO) has led several initiatives towards the elimination of cervical cancer as a public health concern [8]. These initiatives have three main targets in order to eliminate cervical cancer by 2030: (i) to vaccinate 90% of young girls against HPV, (ii) to screen 70% of the target population with a highly sensitive test by age 35 and again by age 45, and (iii) to provide appropriate management to 90% of women identified with cervical disease [9].

One option recommended by the WHO’s to prevent cervical cancer in low-income countries is HPV primary screening followed by triage of HPV-positive women using visual assessment [8]. However, interpreting visual inspection with acetic acid and lugol’s iodine (VIA/VILI) with the naked eye alone is subjective and can be highly variable between health care providers, an issue that a supportive approach with the use of digital image capture can help to overcome [10-13]. To mitigate some of the weaknesses in successful implementation of VIA, investigators have developed enhanced digital imaging of the cervix (termed cervicography) with a camera in Zambia [14-16]. The method requiring a camera connected with a television screen allowed an improved assessment of VIA [17, 18].

The advent of smartphones with high-performance cameras represents an alternative to colposcopy and an opportunity for innovative clinical methods which can be implemented with rather inexpensive tools. Smartphones allow digital image acquisition and inspection with acetic acid and lugol’s iodine (hereafter referred as D-VIA/VILI) allowing to reveal cervical features which may not be visible with the unaided eye (Fig 1). Smartphone cameras provide an alternative to colposcopy and a reliable option for integrating telemedicine but, to date, it is still undetermined whether a combined approach of naked-eye and digital VIA/VILI performs better than naked-eye examination for the diagnosis of cervical intraepithelial neoplasia (CIN) grade 2 or worse (CIN2+).Sensitivity of D-VIA/VILI varies between studies, partly due to the diagnostic challenge of CIN2+ lesions and the absence of a valid reference standard. Our hypothesis was that the combined approach (naked-eye VIA/VILI and D-VIA/VILI) compared to conventional naked-eye VIA assessment would lead to improvement of CIN2+ detection. The objective of this study was to evaluate the diagnostic accuracy of CIN2+ detection using a combined approach (naked-eye and digital VIA) as compared to a traditional naked-eye approach.

Fig 1

Digital visualization of a healthy (1) and pathological (2) cervix using acetic acid (VIA) and Lugol’s iodine (VILI).

Material and methods

Setting and study population

This study was nested in a larger cervical cancer prevention project established 20 years ago in Cameroon with the aim to improve cervical cancer prevention [19]. The study protocol has already been published [20, 21]. The study was conducted in Dschang (West Cameroon) between February 2019 and March 2020. Women aged 30 to 49 years old living in the health district area were invited to participate in a free cervical cancer screening campaign. Exclusion criteria were a positive history of cervical cancer, cervical cancer treatment or hysterectomy, and pregnancy at the time of enrolment.

Procedure

The screening and (if needed) treatment process took place over one day, as recommended by the WHO (Fig 2) [8]. After providing signed informed consent, participants were administered questionnaires by a midwife on their socio-demographic characteristics, gynecological and obstetrical history, smoking status and HIV status. Participants were then invited to perform HPV self-sampling (Self-HPV) as primary screening followed by on-site analysis using the Xpert HPV assay (GeneXpert; Cepheid, Sunnyvale, CA, USA). In case of a negative HPV test, women were advised to repeat screening in 5 years. If the HPV test was positive, a pelvic exam was performed with a PAP-test, followed by triage by VIA/VILI, and finally sampling by endocervical brushing and a biopsy at the site of pathological VIA/VILI or at 6 o’clock when VIA/VILI was negative. The cervical exam was performed by certified midwives. Midwives independently performed a naked-eye cervical evaluation: native (i.e., product-free visualization of the cervix), after application of acetic acid and Lugol’s iodine. A second midwife independently captured images during the pelvic exam (one native, one after application of acetic acid and one after application of Lugol’s iodine) using a smartphone (Samsung Galaxy J5, Seoul, Korea). The smartphone allowed more time for a detailed inspection of the photo, to magnify the lesion by zooming in, and to compare native, acetic acid and iodine colorations by sliding through pictures, thus potentially improving detection of abnormalities.

Fig 2

Flowchart of the study.

* if one of the two tests (VIA or D-VIA) was positive, the decision was to treat. Abbreviations: Self HPV = HPV self-sampling, NA = not analysable, VIA = visual inspection of the cervix with acetic acid; VILI = visual inspection of the cervix with Lugol’s iodine.

VIA/VILI and D-VIA/VILI were independently interpreted as non-pathological (no precancerous lesion) or pathological (suspicious of a precancerous or cancerous lesion). Decision to treat was based on the combined results (VIA/VILI and D-VIA/VILI); if one of the two tests (VIA/VILI or D-VIA/VILI) was positive, the decision was to treat. If the final decision was to treat, women were immediately treated by thermal ablation if eligible, or referred for further evaluation if non-eligible for thermal ablation.

Smartphone photography allowed easy storage and retrieval of VIA images for quality control and educational purposes. Supervision of the screening activity was performed twice a month by a gynecologist specialized in colposcopy who reviewed all photos with the frontline health care providers. Patients who were undertreated according to the specialist’s diagnosis, as well as those with CIN2+ on histological analysis which were not identified through VIA/VILI or D-VIA/VILI, were called back for treatment. If the final decision was not to treat, women were invited for a follow-up in 1 year. The study data collection was conducted by midwives under the supervision of an on-site research assistant through standardized case report forms, and later transcribed to a web-based electronic database (secuTRial®). Smartphone images were anonymously stored on the server of the Geneva University Hospitals in a password-protected computer.

VIA/VILI interpretation

A lower threshold for VIA positivity than is usually recommended by the International Agency for Research on Cancer criteria was applied [22]. ABCD criteria as previously reported were used for diagnosis [23]. Briefly, we considered any acetowhite lesion (faint, translucent, or dense) larger than 5 mm touching the cervical transformation zone, including those that were indeterminate or uncertain, to be positive (or pathological). We considered any lesion revealed by iodine that “matched” an acetowhite lesion to be a positive VILI.

Reference standard

A cervical biopsy and endocervical curettage (ECC) were performed on all participants having a positive HPV test. This material was analyzed at the University Hospitals of Geneva pathology laboratory (the pathology laboratory at Hospital of Gynecology, Obstetrics and Pediatrics of Yaounde was temporary closed due to renovations). Histological results, used as the reference standard, were classified by cervical intraepithelial neoplasm grade. A negative result was considered a cervix without CIN or a CIN1. Positive results included CIN2 or worse, including CIN2, CIN3, adenocarcinoma in-situ (AIS) and cancer.

Ethical considerations

The protocol has obtained approval from the Cantonal Ethics Board of Geneva, Switzerland (Commission cantonale d’éthique de la recherche, N°2017–01110 and N°2020–00868) and the National Ethics Committee for Research on Human Health, Cameroun (Comité national d’éthique de la recherche pour la santé humaine, CNERSH, N°2018/07/1083/CE/CNERSH/SP). The study protocol was registered under ClinicalTrials.gov (number NCT03757299). All participants provided informed written consent before enrolment in the study.

Statistical analysis

We estimated the minimum sample size required based on an expected difference of sensitivity of 20% between the two methods, with a power of 80% to detect it and a two-sided level of significance of 0.05. Considering an estimated 20% of positive HPV tests and 10% of CIN2+ lesions among HPV-positive women (estimated from our previous experiences in Cameroon), we planned to recruit 1,500 participant, using the formula n = (Zα/2+Zβ)2 * (p1(1-p1)+p2(1-p2)) / (p1-p2)2, where n is the sample size, Zα/2 is the critical value of the normal distribution at α/2, Zβ is the critical value of the normal distribution at β and p1 and p2 are the expected sample proportions of the two groups. However, due the COVID-19 pandemic, screening activities at the clinical site were suspended in March 2020. Therefore, we had to consider this analysis with a smaller than expected sample size. The sensitivity and specificity of each method (naked-eye VIA/VILI vs combined naked-eye and D-VIA/VILI) were calculated based on intention to treat, using the histological results as the reference standard. The positivity of each method was also determined as well as positive and negative predictive values (PPV and NPV). P-values comparing sensitivities and specificities between methods were estimated using McNemar’s Chi2 test, and p-values for NPV and PPV using a generalized score statistic for paired designs [24].

Results

Participants’ characteristics

During the period under study, 1081 women (median age 40, IQR 35–45) were recruited for HPV-sampling, with an HPV positivity rate of 17.4%. One hundred and eighty-eight HPV-positive women underwent a pelvic exam for visual assessment. The HPV genotypes identified were HPV-16 (4.3%), HPV-18/45 (8.0%), and other high-risk HPV types pooled together (77.1%). Eighteen women were infected with more than one HPV, including HPV-16 and 18 (1.6%), 16 and other (4.3%), or 18 and other (3.7%). No woman had 3 different types of HPV. Table 1 reports the participants’ sociodemographic and clinical data.

Table 1

Baseline sociodemographic characteristics and reproductive health indicators.

Variable	Total
Participants recruited, n	1081
HPV status, n (%)
Positive	188 (17.4)
Negative	893 (82.6)
Age (years), median (IQR)	40 (35–45)
Marital status, n (%)
Single	87 (8.1)
With partner	449 (41.6)
Married	458 (42.5)
Divorced/widowed	85 (7.9)
Education, n (%)
Unschooled	7 (0.7)
Primary education	320 (29.7)
Secondary education	580 (53.8)
Tertiary education	171 (15.9)
Other	1 (0.1)
Employment status, n (%)
Employed without responsibility	239 (22.2)
Employed with responsibility	34 (3.2)
Self-employed	296 (27.4)
Housewife	244 (22.6)
Farmer	226 (21.0)
Other	40 (3.7)
Age at menarche (years), mean ± SD	14.7 (±1.8)
Age at first intercourse, median (IQR)	18 (16–19)
Number of sexual partners, median (IQR)	3 (2–5)
Contraception, n (%)
None	786 (72.7)
Condom	86 (8.0)
Hormonal contraception	184 (17.0)
Other	8 (0.7)
Unknown	4 (0.4)
HIV status, n (%)
Negative	1019 (94.3)
Positive	36 (3.3)
Unknown	26 (2.4)
Previous HPV infection, n (%)
Negative	901 (83.4)
Positive	178 (16.5)
Treated	23 (2.1)
Unknown	2 (0.2)
Age at first delivery (years), mean ± SD	21.0 (±5.2)
Parity, n (%)
Nulliparous	44 (4.1)
1–4	467 (43.3)
>4	568 (52.6)
Gravidity, n (%)
Nulligravida	25 (2.3)
1–4	315 (29.2)
>4	739 (68.5)
Tobacco
Non-smoking	1057 (98.1)
Smoking	21 (1.9)

Abbreviations: IQR = interquartile range; SD = standard deviation; HIV = human immunodeficiency virus; HPV = human papillomavirus.

Histological results

Among 188 biopsies, 26 (13.8%) were pathological (CIN2+), 155 non-pathological (82.4%), which were either negative or CIN1, and 7 (3.7%) could not be analyzed because of insufficient material. In these 26 pathological lesions, there were 11 CIN2, 14 CIN3 and one cancerous lesion. Table 2 shows the results of visual assessment according to histology.

Table 2

VIA/VILI, D-VIA/VILI and combined assessment of histopathologic results, categorized by CIN grade.

CIN grade n (%)	VIA/VILI		D-VIA/VILI		COMBINED VIA/VILI & D-VIA/VILI
	Positive	Negative	Positive	Negative	Positive	Negative
Negative*	73 (64.6)	40 (35.4)	72 (63.2)	42 (36.8)	82 (71.9)	32 (28.1)
CIN1	33 (80.5)	8 (19.5)	34 (82.9)	7 (17.1)	37 (90.2)	4 (9.8)
CIN2**	9 (81.8)	2 (18.2)	9 (90.0)	1 (10.0)	10 (90.9)	1 (9.1)
CIN3	11 (78.6)	3 (21.4)	13 (92.6)	1 (7.1)	13 (92.6)	1 (7.1)
Cancer	1 (100)	0	1 (100)	0	1 (100)	0
Total CIN2+	21 (80.8)	5 (19.2)	23 (92.0)	2 (8.0)	24 (92.3)	2 (7.7)

Abbreviations: CIN = cervical intraepithelial neoplasm; VIA = visual inspection of the cervix with acetic acid; VILI = visual inspection of the cervix with Lugol’s iodine; DVIA = digital visual inspection of the cervix with acetic acid; DVILI = digital visual inspection of the cervix with Lugol’s iodine.

*1 missing data for VIA/VILI.

**1 non-interpretable D-VIA/VILI.

VIA/VILI and D-VIA/VILI diagnostic accuracy

If either VIA/VILI or D-VIA/VILI was positive the combined visual assessment was considered positive. Note that patient 11 in the Table is reported as positive after VIA/VILI assessment but was not treated as the D-VIA/VILI was of insufficient quality for assessment, due to the presence of blood. D-VIA/VILI identified three patients with precancerous lesions that VIA/VILI had missed. Both patients with CIN2+ identified by histological analysis who were missed by VIA and D-VIA were called back and subsequently treated. Out of 188 participants, there were 187 results of VIA/VILI and 187 of D-VIA/VILI. VIA’s positivity was 69.4% and that of D-VIA/VILI 72.4%. VIA/VILI’s sensitivity was 80.8% (95% CI 60.6–93.4), specificity was 31.2% (95% CI 24–39.1), PPV was 16.5% (95% CI 10.5–24.2) and NPV was 90.6% (95% CI 79.3–96.9). For D-VIA/VILI, sensitivity was 92.0% (95% CI 74.0–99.0, p = 0.833 with VIA/VILI as reference group), specificity was 31.6% (95% CI 24.4–39.6, p = 0.8415), PPV was 17.8% (95% CI 11.7–25.5, p = 0.3268) and NPV was 96.1% (95% CI 86.5–99.5, p = 0.0771). As for the combination of VIA/VILI and D-VIA, sensitivity was 92.3% (95% CI 74.9–99.1, p = 0.0833 with VIA/VILI as reference group), specificity 23.2% (95% CI 16.8–30.7, p = 0.0005), PPV 16.8% (95% CI 11.1–23.9, p = 0.7500) and NPV 94.7% (95% CI 82.3–99.4, p = 0.1963). Table 3 summarizes these results. Considering our final sample of 26 CIN2+ participants, the current study was powered at 43.6% for the detection of a 20% difference between the sensitivities of VIA/VILI and the combined VIA/VILI and D-VIA/VILI method.

Table 3

Sensitivity, specificity, PPV and NPV of each diagnostic method.

Diagnostic Method	Sensitivity % (95% CI)	Sensitivity’s p-value (vs VIA/VILI)	Specificity % (95% CI)	Specificity’s p-value (vs VIA/VILI)	PPV % (95% CI)	PPV’s p-value (vs VIA/VILI)	NPV % (95% CI)	NPV’s p-value (vs VIA/VILI)
VIA/VILI	80.8 (60.6–93.4)		31.2 (24–39.1)		16.5 (10.5–24.2)		90.6 (79.3–96.9)
D-VIA/D-VILI	92.0 (74.0–99.0)	0.0833	31.6 (24.4–39.6)	0.8415	17.8 (11.7–25.5)	0.3268	96.1 (86.5–99.5)	0.0771
VIA/VILI and D-VIA/D-VILI combined	92.3 (74.9–99.1)	0.0833	23.2 (16.8–30.7)	0.0005	16.8 (11.1–23.9)	0.7500	94.7 (82.3–99.4)	0.1963

Abbreviations: VIA = visual inspection of the cervix with acetic acid; VILI = visual inspection of the cervix with Lugol’s iodine; PPV = positive predictive value; NPV = negative predictive value; D-VIA = digital visual inspection of the cervix with acetic acid; D-VILI = digital visual inspection of the cervix with Lugol’s iodine.

Discussion

Triaging of HPV-positive women with cytology or biomarkers is not readily implementable in low-resource settings. Therefore, alternative methods like VIA have been considered. Some reports found that VIA used as a triage method is associated with an important loss of sensitivity and may be inappropriate for use in a triaging strategy [25, 26]. The main reason is that interpreting VIA by naked eye alone is a highly subjective method and widely depends on the experience of the provider [27-29]. In this project, we evaluated if acquisition of digital cervical images (native, after VIA and VILI) with a smartphone has the potential to improve traditional unaided cervical visual assessment to detect precancerous lesions. Camera or smartphone-enhanced VIA/VILI represent an innovative method and a pragmatic alternative to colposcopy adapted to low-resource contexts [18]. However, the major question is to determine if D-VIA/VILI has an impact on the clinical management of HPV-positive women as compared to naked-eye evaluation. A critical issue for the patient and the provider is, first of all, to reduce the risk of missing a CIN2+ lesion, rather than improving diagnostic accuracy.

To the best of our knowledge, this is the first prospective trial which aimed to investigate the impact of the adjunct of D-VIA/VILI to conventional naked-eye examination. Previous studies on the possible role of camera-enhanced diagnosis of cervical precancerous lesions and monitoring with digital equipment have been reported, with sensitivity and specificity of D-VIA ranging between 66.7% and 94.1%, and 50.4% and 85.4% respectively [17, 30], which is consistent with our results. However, little is known about the addition of digital images to the final decision regarding precancerous lesion management in low-resource settings [10, 17].

Our study supports that the addition of smartphone-enhanced VIA/VILI may lead to improvement in CIN2+ detection, although only borderline statistical significance was reached in regard to sensitivity. Considering D-VIA/VILI individually, we also observed a trend in sensitivity improvement. Its positive and negative predictive values were also slightly higher, however without reaching statistical significance. On the other hand, digital assessment reduced specificity compared to the naked eye alone. A loss in specificity may however be considered acceptable in low-resource contexts where loss to follow-up is a common issue, and considering the availability of safe and well-tolerated therapeutic options such as thermal ablation. The combined use of VIA/VILI and D-VIA/VILI further allowed to detect two patients with a CIN3 lesion, and one patient with a CIN2 lesion.

The inclusion of a digital approach to cervical screening strategies could be a good adjunct to VIA/VILI. The combination of a conventional naked-eye approach with a digital approach has already been performed in dermatology for the detection of melanomas and showed an improvement in performance [31]. Although not a scientific validation of the technique, in daily practice, we observed that front-line providers having experience with D-VIA/VILI acquisition unanimously prefer the use of digital images, simply because it offers the possibility to be manipulated to zoom in on suspicious regions or transformation zones, as well as the possibility to simultaneously compare native, VIA and VILI images and ask colleagues for a second opinion in real time. In cases of doubt, D-VIA/VILI could be verified remotely by a more experienced person [32].

The main limitation of the study is the sample size which was inferior to what was planned because of the COVID-19 pandemic, which required to temporarily shut down the screening unit and, as a consequence, to terminate the study prematurely. A strength of our study is that it included only HPV-positive women in routine clinical practice. Second, the use of systematic biopsies and endocervical curettage for all HPV-positive women as standard reference regardless of the visual assessment interpretation avoided verification bias.

Replication with a larger sample size would be necessary to draw definitive conclusions. Nonetheless, most providers are convinced that the digital cervical image is useful for the diagnosis of CIN2+ lesions. Although the primary endpoint was inconclusive, this study represents the best available evidence to date that suggests that D-VIA/VILI may potentially improve cervical cancer screening.

Other avenues for improving cervical cancer screening exist, and new strategies to improve accuracy of visual assessment need to be investigated. The development of a smartphone application recognizing pathological lesions of the cervix, for example, is underway, and might contribute to improving the screening process [33].

Conclusion

A digital approach performed in addition to naked-eye evaluation may be of clinically relevant benefit in CIN2+ detection. Indeed, the combination of VIA/VILI and D-VIA/VILI seems to provide an increase in sensitivity, with an acceptable decrease in specificity. Digital assessment is an innovative tool that may contribute to changing the current management of cervical cancer prevention in low-income countries. However, this will need to be confirmed by further studies with a larger sample size.

7 Oct 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Michael Scheurer, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Addition of digital VIA to conventional naked-eye examination for triage of HPV-positive women: A study conducted in a low-resource setting

Thank you for the opportunity to review this paper. It describes a study conducted in West Cameroon to evaluate the improvement in diagnostic performance when supplementing naked-eyed VIA/VILI with digital VIA/VILI using a low-cost smartphone camera. This low cost technology is exciting as it could potentially improve test-see-treat programs in low-resource settings by providing an inexpensive and readily accessible tool that may improve the sensitivity of VIA/VILI for triage of high-risk HPV positive patients. I have a couple of comments/suggestions for the authors:

1) Table 2 lists the VIA/VILI and D-VIA/D-VILI results of each of the 26 patients with CIN2+, stratified by CIN grade. The table is detailed in providing data at the participant level. However, it would be helpful to summarize the data for the reader. Would it be possible to present CIN grade- stratified 2x2 tables, indicating the number (%) within each cell? This would additionally provide the benefit of readily showing the concordance of naked VIA and D-VIA.

VIA vs. D-VIA 2x2

CIN2

VIA vs. D-VIA 2x2

CIN3

VIA vs D-VIA 2x2

Cancer

VIA vs D-VIA 2x2

2) Is there any reason why data for patients with low grade CIN aren’t provided? Following the format suggested above, it would be helpful to have a 2x2 table for patients with

3) Consort diagram and lines 210-213 on page 13. The authors indicate that 1 patient had a D-VIA with insufficient quality for assessment. However, the next sentence indicates that 188/188 patients with CIN2+ had D-VIA results, while 186 had VIA results.

4) It may be worth reporting on statistically significant differences in negative predictive value—one of the values of good screening/triage tests is providing reassurance to patients with negative triage tests that they can return in one year without fearing that they were mistakenly under-managed. While I didn’t do the math, it seems like a jump from 90.6% NPV to 94.7% is a significant improvement in the reassurance rate.

5) As the target sample size was understandably not attained due to COVID-19, it would be worth mentioning the power of the study given the actual sample size.

6) A few minor editorial comments:

a. Page 6, Procedure—“Midwives independently performed a naked-eye cervical evaluation (native, after acetic acid and Lugol’s iodine). Can you use i.e., after you use the word native to define what that means?

b. Page 4, Introduction, Line 74-76: “Advances in technology for ….. are adapted to the context.” It is unclear what is meant by “adapted to the context” (i.e., are contextual adaptations based on capacity and availability of resources/infrastructure?)

Reviewer #2: Summary: This study tested how well digital-VIA with a smartphone aided in diagnosis of cervical dysplasia when added to naked-eye VIA. These tests were compared to histological diagnosis of CIN2. The study had to be cut short because of COVID but provided some interesting data that suggests that D-VIA helped in diagnosis and same-day treatment of cervical dysplasia.

Major comments:

1. The idea of using a smartphone with digital images of the cervix to aid in diagnosis has the potential for a low cost and innovative improvement to traditional VIA.

2. The manuscript should be reviewed by a native-English speaker and editor as there are numerous sentences that could be written more clearly and spelling/grammatical mistakes could be corrected.

3. Who reviewed the VIA and D-VIA to determine if they needed treatment? Were these midwives? What if there was a difference of opinion between providers?

4. The methods section in regards to procedures needs to be expanded significantly: specifically, what were the questionnaires asking participants? When was biopsy and ECC being performed? Was this before or after VIA? Did everyone get VIA and VILI? How many images were taken? How were these stored or disposed of securely?

5. Table 1 only includes those who are HPV positive (188) but I think this really should be the total women recruited of 1081 and their characteristics.

6. Are the WHO recommendations for treating dysplasia only if CIN2+? Of those who were not treated as VIA and D-VIA missed dysplasia, how many were called back and successfully treated?

7. I would try and provide some more structure to the introduction: what is known about cervical cancer and its screening and treatment; what is not known; and then what you are hoping to accomplish. Some of the details of VIA and D-VIA could be included in the methods as your ‘intervention’.

8. Was regular VIA also completed prior to digital VIA? I imagine there could be some bias if digital VIA was done before naked-eye VIA.

Minor comments:

1. VIA is visual inspection with acetic acid and VILI is visual inspection with Lugol’s iodine. VILI has different test characteristics than VIA and so technically VIA should just include discussion of acetic acid and not also VILI.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

13 Mar 2022

Reviewer #1: Addition of digital VIA to conventional naked-eye examination for triage of HPV-positive women: A study conducted in a low-resource setting

Thank you for the opportunity to review this paper. It describes a study conducted in West Cameroon to evaluate the improvement in diagnostic performance when supplementing naked-eyed VIA/VILI with digital VIA/VILI using a low-cost smartphone camera. This low cost technology is exciting as it could potentially improve test-see-treat programs in low-resource settings by providing an inexpensive and readily accessible tool that may improve the sensitivity of VIA/VILI for triage of high-risk HPV positive patients. I have a couple of comments/suggestions for the authors:

1) Table 2 lists the VIA/VILI and D-VIA/D-VILI results of each of the 26 patients with CIN2+, stratified by CIN grade. The table is detailed in providing data at the participant level. However, it would be helpful to summarize the data for the reader. Would it be possible to present CIN grade- stratified 2x2 tables, indicating the number (%) within each cell? This would additionally provide the benefit of readily showing the concordance of naked VIA and D-VIA.

Thank you for this suggestion. We have summarized Table 2 according to your recommendations.

Table 2. VIA/VILI, D-VIA/VILI and combined assessment of histopathologic results, categorized by CIN grade

CIN grade

2) Is there any reason why data for patients with low grade CIN aren’t provided? Following the format suggested above, it would be helpful to have a 2x2 table for patients with

We have added VIA/VILI and D-VIA/VILI assessment for negative and CIN1 cases in the table above.

3) Consort diagram and lines 210-213 on page 13. The authors indicate that 1 patient had a D-VIA with insufficient quality for assessment. However, the next sentence indicates that 188/188 patients with CIN2+ had D-VIA results, while 186 had VIA results.

We agree with this comment and there was indeed some confusion in these numbers. We have now excluded the non-interpretable case from the final analysis for D-VIA/VILI. In doing so, the sensitivity and negative predictive value of D-VIA/VILI have slightly changed. Changes have been made in tables 2 and 3, as well as in the manuscript.

4) It may be worth reporting on statistically significant differences in negative predictive value—one of the values of good screening/triage tests is providing reassurance to patients with negative triage tests that they can return in one year without fearing that they were mistakenly under-managed. While I didn’t do the math, it seems like a jump from 90.6% NPV to 94.7% is a significant improvement in the reassurance rate.

Thank you for this comment. P-values have been added for the difference in PPV and NPV between VIA/VILI and D-VIA/VILI and the combined test respectively, but showed no statistical significance at the p=0.05 level.

5) As the target sample size was understandably not attained due to COVID-19, it would be worth mentioning the power of the study given the actual sample size.

Thank you for this comment. We have added the current study power in the results section.

Lines 238: Considering our final sample of 26 CIN2+ participants, the current study was powered at 43.6% for the detection of a 20% difference between the sensitivities of VIA/VILI and the combined VIA/VILI and D-VIA/VILI method.

6) A few minor editorial comments:

a. Page 6, Procedure—“Midwives independently performed a naked-eye cervical evaluation (native, after acetic acid and Lugol’s iodine). Can you use i.e., after you use the word native to define what that means?

The following has been specified:

Lines 131: Midwives independently performed a naked-eye cervical evaluation: native (i.e., product-free visualization of the cervix), after acetic acid and Lugol’s iodine.

b. Page 4, Introduction, Line 74-76: “Advances in technology for ….. are adapted to the context.” It is unclear what is meant by “adapted to the context” (i.e., are contextual adaptations based on capacity and availability of resources/infrastructure?)

We have completed this paragraph as follows:

Lines 74: Advances in technology for cervical detection in low-resource settings using HPV primary testing followed by immediate treatment with cryotherapy or thermal ablation if needed are adapted to the context (5)(6)(7). Indeed, a ‘screen-and-treat’ approach in a single visit requires fewer resources and reduces risk of loss to follow-up.

Reviewer #2: Summary: This study tested how well digital-VIA with a smartphone aided in diagnosis of cervical dysplasia when added to naked-eye VIA. These tests were compared to histological diagnosis of CIN2. The study had to be cut short because of COVID but provided some interesting data that suggests that D-VIA helped in diagnosis and same-day treatment of cervical dysplasia.

Major comments:

1. The idea of using a smartphone with digital images of the cervix to aid in diagnosis has the potential for a low cost and innovative improvement to traditional VIA.

We thank the reviewer for this comment.

2. The manuscript should be reviewed by a native-English speaker and editor as there are numerous sentences that could be written more clearly and spelling/grammatical mistakes could be corrected.

The manuscript has been reviewed and the grammatical mistakes have been corrected.

3. Who reviewed the VIA and D-VIA to determine if they needed treatment? Were these midwives? What if there was a difference of opinion between providers?

During the examination, one midwife independently assessed the need for treatment based on VIA/VILI and a second midwife independently based on D-VIA/VILI. A research assistant working on site ensured the quality of the data collected. In cases of disagreement, one positive assessment (either VIA/VILI or D-VIA/VILI) was sufficient to provide treatment. A gynecologist specialized in colposcopy reviewed all photos twice a month and gave her diagnosis based on D-VIA/VILI. Patients who were undertreated according to the specialist’s diagnosis were recalled for treatment.

The methods section has been completed accordingly:

Lines 131: Midwives independently performed a naked-eye cervical evaluation: native (i.e., product-free visualization of the cervix), after application of acetic acid and Lugol’s iodine. A second midwife independently captured images (native, after acetic acid and Lugol’s iodine) using a smartphone (Samsung Galaxy J5, Seoul, Korea). VIA and D-VIA were independently interpreted as non-pathological (no precancerous lesion) or pathological (suspicious of a precancerous or cancerous lesion). Decision to treat was based on the combined results (VIA and D-VIA); if one of the two tests (VIA or D-VIA) was positive, the decision was to treat. If the final decision was to treat, women were immediately treated by thermal ablation if eligible, or referred for further evaluation if non-eligible for thermal ablation.

Smartphone photography allowed easy storage and retrieval of VIA images for quality control and educational purposes. Supervision of the screening activity was performed twice a month by a gynecologist specialized in colposcopy who reviewed all photos with the frontline health care providers. Patients who were undertreated according to the specialist’s diagnosis, as well as those with CIN2+ on histological analysis which were not identified through VIA/VILI or D-VIA/VILI, were called back for treatment. If the final decision was not to treat, women were invited for a follow-up in 1 year.

4. The methods section in regards to procedures needs to be expanded significantly: specifically, what were the questionnaires asking participants? When was biopsy and ECC being performed? Was this before or after VIA? Did everyone get VIA and VILI? How many images were taken? How were these stored or disposed of securely?

Thank you for this comment. The methods section has been completed with the following information.

Lines 123: After providing and signed informed consent, participants were administered questionnaires by a midwife on their socio-demographic characteristics, gynecological and obstetrical history, smoking status and HIV status.

Lines 128: If the HPV test was positive, a pelvic exam was performed with a PAP-test, followed by triage by VIA/VILI, and finally sampling by endocervical brushing and a biopsy at the site of pathological VIA/VILI or at 6 o’clock when VIA/VILI was negative.

Lines 133: A second midwife independently captured images during the pelvic exam (one native, one after application of acetic acid and one after application of Lugol’s iodine) using a smartphone (Samsung Galaxy J5, Seoul, Korea).

Lines 153: Smartphone images were anonymously stored on the server of the Geneva University Hospitals in a password-protected computer.

5. Table 1 only includes those who are HPV positive (188) but I think this really should be the total women recruited of 1081 and their characteristics.

Changes to Table 1 have been made according to the suggestion.

6. Are the WHO recommendations for treating dysplasia only if CIN2+? Of those who were not treated as VIA and D-VIA missed dysplasia, how many were called back and successfully treated?

Indeed, when histological results are available, the WHO recommends to treat only women with CIN2+ (WHO guidelines for treatment of cervical intraepithelial neoplasia 2-3 and adenocarcinoma in situ, 2014). In our study, all untreated participants in which CIN2+ was identified through histological analysis were called back and successfully treated. This has been added to the manuscript.

Methods - lines 148: Patients who were undertreated according to the specialist’s diagnosis, as well as those with CIN2+ on histological analysis which were not identified through VIA/VIL or D-VIA/VILI, were called back for treatment.

Results – lines 227: Both patients with CIN2+ identified by histological analysis who were missed by VIA and D-VIA were called back and subsequently treated.

7. I would try and provide some more structure to the introduction: what is known about cervical cancer and its screening and treatment; what is not known; and then what you are hoping to accomplish. Some of the details of VIA and D-VIA could be included in the methods as your ‘intervention’.

Thank you for this comment. We have restructured the introduction accordingly.

8. Was regular VIA also completed prior to digital VIA? I imagine there could be some bias if digital VIA was done before naked-eye VIA.

Regular VIA was performed independently by the first midwife. Pictures of the cervix were captured by the second midwife before application of acetic acid (native) and 60 seconds after application of acetic acid (VIA), which were not shown to the first midwife prior to her assessment. Lugol’s iodine was then applied for naked-eye assessment by the first midwife, before being photographed by the second midwife for independent digital assessment.

Minor comments:

1. VIA is visual inspection with acetic acid and VILI is visual inspection with Lugol’s iodine. VILI has different test characteristics than VIA and so technically VIA should just include discussion of acetic acid and not also VILI.

Thank you for this comment. VILI has been added throughout the manuscript when

appropriate.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

21 Apr 2022

Addition of digital VIA to conventional naked-eye examination for triage of HPV-positive women: A study conducted in a low-resource setting

PONE-D-21-15964R1

Dear Dr. Dufeil,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Michael Scheurer, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors have done a very nice job in addressing the reviewer comments. I just have a few minor comments but otherwise feel this manuscript should be accepted.

Minor comments:

1. In abstract, would not switch between presenting raw numbers or percentages first. I would pick one and use it continuously through the manuscript.

2. Second paragraph of introduction, first sentence could read for better clarity: “Cervical cancer screening allows for the detection and prevention of disease a precancerous stage…”

3. In discussion, the main limitation sentence should read: “The main limitation of the study…which required a temporary shut down of the screening unit, and as a consequence, to terminate the study prematurely.”

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

4 May 2022

PONE-D-21-15964R1

Addition of digital VIA/VILI to conventional naked-eye examination for triage of HPV-positive women: A study conducted in a low-resource setting

Dear Dr. Dufeil:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Michael Scheurer

Academic Editor

PLOS ONE