跳到正文
原文
Nature 健康护理主题·· 18 天前精选AI 评分65

印度资源有限地区利用多重生物标志物的宫颈癌现场筛查设备

Cervical cancer on-site mass-screening device utilizing multiple biomarkers in Swab samples for surveillance in the resource limited areas of India

AI 导读

研究团队开发了一种便携式宫颈癌筛查设备KayaCervi,通过多标记组合提高了筛查的准确性和效率。该设备在印度资源有限地区进行了社区筛查,显示了98%的高特异性和38.5%的阳性预测值。

推荐理由

研究团队开发了一种便携式宫颈癌筛查设备,通过多标记组合提高了筛查的准确性和效率,适用于资源有限地区的初步筛查。

正文

Introduction

The World Health Assembly adopted a strategy in August 2020 to eliminate cervical cancer by 2030, establishing targets of 90% HPV vaccination, 70% screening, and 90% treatment, with the aim of lowering the incidence to fewer than 4 cases per 100,000 women. Human papillomavirus (HPV) is the primary cause of nearly 95% of cervical cancer cases. The cervix is the lower part of the uterus that opens into the vagina (birth canal). Persistent HPV infection can induce precancerous changes in cervical cells, which, if left undetected and untreated, may gradually progress to invasive cervical cancer. In immunocompetent women, this progression typically takes 15–20 years. However, in women with compromised immunity, such as those with untreated HIV infection, the disease may progress more rapidly, within 5–10 years. Several factors influence the risk of progression, including the oncogenic potential of the HPV genotype, immune status, coexisting sexually transmitted infections, high parity, early age at first pregnancy, prolonged use of hormonal contraceptives, and tobacco smoking (Stelze, Dominik et al. 2020)1,2). Because of the limitations associated with advanced diagnostics, as well as the expense and time associated with procedures such as Pap smears3,4, efficient and affordable devices that can be used in rural India are critical in terms of detecting the problem early, providing follow-ups, and improving outcomes for patients. The presented technology addresses one of the reasons for high mortality rates among Indian women caused by poor hygiene, lack of knowledge, and inaccessible screening, although vaccines are available5,6. Pap smear is a quick, painless, and minimally invasive examination with the sensitivity of 51% and the specificity of 67%. It has been an essential procedure in the whole world for 50 years. Nevertheless, the wide usage in rural areas is impossible due to limitations associated with urban hospitals, expenses, low sensitivity, high frequency of false negatives, and the necessity of several visits leading to delayed diagnosis7,8,9. On the other hand, the gold standard, cervical biopsies, is an invasive procedure requiring a laboratory, but having high sensitivity and specificity10. Key biomarkers in cervical neoplasms include Ki-67, cyclin-dependent kinases, and HPV infection markers such as p16INK4a, which indicates active HPV E7 expression. p16 inhibits cyclin-dependent kinase CDK4/6 and, when overexpressed, signifies active HPV E7 oncogene expression and retinoblastoma gene suppression, commonly seen in HPV-driven tumors. HPV DNA is detected in 94% of in situ lesions and 76–85% of adenocarcinomas. Overexpression of p16, identified through immunohistochemistry (IHC), serves as a reliable marker for diagnosing both squamous and glandular cervical neoplasms. Despite its utility, there is limited data on biomarker testing using cervix swab samples for early detection in mass screening for cervical cancer.

Thus, ten biomarkers p16INK4a, Ki67, HPV-16, HPV 18, MCM5, CA-125, CA19-9, ACTN-4, CEA, CD146 were majorly evaluated in the collected cervix swab samples to see if they are on a causal pathway in the disease pathogenesis indicative of signs of tumorigenesis with high sensitivity and specificity comparable to the existing Pap-smear tests. The prototype device KayaCervi validations using biomarkers selected and analyzed by ELISA test was done. A mass-screening device for cervical cancer was developed using, technology from Kaya-17 USA. It was repurposed to create a prototype capable of detecting potential cervical cancer biomarkers in a single test. This technology was chosen for its robust platform, which does not require advanced infrastructure and is highly reproducible (US Army Project, 2007, 2008)11. The device, patented under U.S. Patent No. 16/690,589 and Indian Patent No. 201941011177, is designed to work with a USB port. It detects cervical cancer antigens expressed during the precancerous stage, providing results in Relative Fluorescence Units (RFU) using in attached reader.

The study compares the expression patterns of all biomarkers in normal, dysplastic, and cancerous cervical tissues, assessing their diagnostic potential through ELISA, and KayaCervi device tests. Simultaneously KayaCervi device reports compared with the Pap smear, LBC and IHC tests to achieve high sensitivity and specificity. Study tested 200 suspected cervical cancer swab samples (OPD) that out of ten biomarkers tested; p16INK4a, Ki67, HPV-16, HPV 18 are ideal biomarker combination that can be used in mass-screening of cervical. These four biomarkers, p16INK4, Ki67, HPV-16, and HPV-18, were likely represent a powerful combination that addresses different aspects of cervical cancer development. HPV-16 and HPV-18 identify the primary cause, p16INK4 indicates active viral transformation, and Ki67 highlights abnormal cell proliferation. This multi-marker approach provides a high level of sensitivity and specificity (> 90%), making it ideal for accurately identifying individuals at risk in a mass-screening setting compared to the other tested biomarkers. As the proposed device is intended for community-based mass screening, and the primary objective of the community-based validation study was to evaluate the performance of the device as a mass screening tool among apparently healthy women. In such settings, the majority of participants are disease-negative, and high specificity is essential to minimize false-positive results, unnecessary referrals, psychological distress, and healthcare costs. Therefore, the validation study focused primarily on the assessment of specificity in a real-world screening population. Sensitivity assessment was undertaken during the device standardization phase using both screening participants and additional confirmed positive cases to ensure adequate representation of disease-positive samples for assay optimization. Therefore, the sample size was determined based on the specificity parameter, resulting in a required sample size of 867 participants from 36 villages from 36 villages ; Bhopal city, Baghmuggalia, Pipaliapende Khan, Misrod, Raisen, AshaPuri, Dhamdusar, Goharganj, Neelbad, Chiklod, Barkhera Pathani, Kalapani, Kolua Kalan, Ratua, Barbatpur, Amoda, Lalpur, Dharna, Chhapra Tola, Ghisee, Ari, Gangapur Tola, Chota Dhrana, Seoni, Nandini, Semra Kalan, Amkheda, Ratibadh, Phanda, Berasia, Bhauri, Bhanpur, Kolar, Salaiya, Huzur, Jatkhedi and Chuna Bhatti where Cervical Cancer awareness, Health Hygiene, camps and samples were collected from the apparently healthy women volunteers (Fig. 1).

Fig. 1

Map: Digital map showing Bhopal and Seoni District where community-based study was conducted.

Materials and methods

Ethical approval

The study was granted ethical approval by the review board (Institutional Human Ethical Committee) of AIIMS Bhopal, India, with the approval number (IHEC LoP– EL071) Extension of LoP for community testing, confirming that it met the necessary ethical guidelines for conducting research involving human subjects under Grant-in-Aid-DHR research grant (R.11016/02/2023-GIA/HR).

The study is as a two-phase translational diagnostic development project:

Phase I (Standardization): Device optimization using screening samples supplemented with confirmed positive cases to establish cut-offs and assess analytical/clinical performance (Fig. 2).

Fig. 2

Flow Chart showing Phase I and II study protocol

Phase II (Community Validation): Real-world evaluation in 865 apparently healthy women undergoing simultaneous LBC testing, with particular emphasis on specificity and feasibility for population-based screening (Fig. 2).

Phase I: Study design for device and biomarkers validation (Hospital Setting)

Study carried out at All India Institute of Medical Sciences, Bhopal, India, during April 2024–March 2025. The inclusion criteria for the study consisted of female (n = 200) outpatients aged 30 to 70 experiencing various gynecological issues, such as vaginal bleeding between periods, heavy or prolonged menstrual bleeding, vaginal discharge, post-coital bleeding, douching, unexplained persistent pelvic and/or back pain, vaginal bleeding after menopause, and symptoms like itching, burning, pain, or tenderness of the vulva. Additionally, patients specifically seeking treatment for cervical complications, including tumors, infections, and inflammation, were enrolled. Exclusion criteria included patients with a history of preoperative chemotherapy, radiotherapy, or malignant tumors, as well as those with prior diagnoses of cervical cancer or related therapeutic interventions. Pregnant and breastfeeding women were also excluded from the study. All sensitivity and specificity estimates reported for the phase I analysis were specifically relative to Pap smear and not to ELISA or individual biomarker positivity.

Phase II: Community based study for device validation

Every month (2024–26) community based study was planned. We selected 36 villages. Initial visits were aims towards community preparations, awareness camps on cervical cancer. During these camps team has interacted roughly 2000 women. In Phase II, KayaCearvi testing and LBC were performed in the community-based cohort. For primary comparison of KayaCervi with cytological findings, LBC was used as the reference cytological method for sensitivity and specificity.

Sample size

We have used a multistage cluster sampling technique to select study participants from the 211 listed villages in Bhopal district. We considered villages as the clusters in our study. We considered only those villages where the expected number of female participants is at least 20. This is to improve operational feasibility of the research process. In the first stage, we have selected the villages using Probability Proportionate to Size (PPS) technique. Then the study team conducted mapping and listing of households with participants in the selected villages. In the second stage of selection, at least 40 eligible participants in the selected villages were recruited. If the selected village has more than 20, but less than 40 participants, the study team included the nearby villages in the selected village as composite cluster to achieve cluster sample size of 40. For sample size calculation, we are considering anticipated sensitivity of 65% and specificity of 70%. The prevalence of pre-malignant and malignant cervical lesions was reported as 6.1% in a previous study12. We have considered 95% confidence level and 20% relative precision for sample size for sensitivity and specificity and used these values in sample size calculation formula for diagnostic test evaluation13.

The formulae for sample size using sensitivity and specificity are:

$${\text{n}}_{{{\text{Se}}}} = \frac{{Z2\frac{\alpha }{2}Se \left( {1 - Se} \right)}}{d2 \times Prevalence}, \, \left[ \begin{gathered} {\text{where}}\;{\text{z}}_{{\alpha /{2}}} \;{\text{is}}\;{\text{z}}\;{\text{score}}\;\left( {{1}.{96}} \right)\;{\text{for}}\;{\text{confidence}}\;{\text{level}}\;\left( {{95}\% } \right), \hfill \\ {\text{S}}_{{\text{e}}} \;{\text{is}}\;{\text{sensitivity}},\;{\text{d}}\;{\text{is}}\;{\text{relative}}\;{\text{precision}} \hfill \\ \end{gathered} \right]$$

$${\text{n}}_{{{\text{Sp}}}} = \frac{{Z2\frac{\alpha }{2}Sp \left( {1 - Sp} \right)}}{{d2 \times \left( {1 - Prevalence} \right)}} \left[ {{\text{where}}\;{\text{S}}_{{\text{p}}} \;{\text{is}}\;{\text{specificity}}} \right]$$

Using the anticipated parameters, the sample size for sensitivity is 865 and for specificity are 45. Since, we using a cluster sampling strategy, we need to control for variation between intracluster and intercluster variation through use of design effect. We also need to control for non-response. Considering design effect of 1.5 and non-response rate of 10%, the final sample size for sensitivity is 1427. Since our cluster size is 40, we need to select 36 villages for our study.

Sample processing

After informed consent, cervical swab samples collected from the test population based on our predefined inclusion and exclusion criteria.

Collected cervical swab sample cell lysate was prepared using a swab saturated with ice-cold PBS solution for 3–4 h. Swab contains cervical epithelial cells, and cells were pellet down at 400xg for 5 min at 40C. After RIPA Lysis Buffer Solution, the pellets sonicated using a Bio-Rad Micro-Roto for cell lysis kit with 50% amplitude14. The supernatant containing proteins collected by passing through the column and subjected to protein estimation using Bicinchoninic Acid (BCA) method described previously by Stoscheck with slight modifications15. This protocol has been submitted for the patent (patent filing No.202321006867). The processed cervical swabs tested by ELISA, on the KayaCervi prototype device along with Pap tests16. PPV, NPV and sensitivity and specificity were calculated. Decanting, PBS and several wash pellets were incubated for 15 min with 100 µl ice-cold. Collected 200 cervix swab samples were process according to the sample processing protocol developed ((filed patent application No 202321006867). ELISA tests conducted on all prospectively collected cervix swab samples using respective ten monoclonal antibodies (Table 1).

Table 1 Name of the monoclonal antibodies use in the current experiments17.

Full size table

Biomarkers used and tested

Cervical cancerous cells produce distinct surface antigens protein p16INK4a, Ki67, HPV-16, HPV 18, MCM5, CA-125, CA19-9, ACTN-4, CEA, CD146 protein, and that its high expression is indicative of early to late EMT stage in the cervix cells. All the cervical swab samples collected from the patients were analyze using a sandwich ELISA test. After standardization of the samples and processing, the protocol was further used for all the biomarkers concentration determination. Out of tested all biomarkers, the p16NK4A, HPV-16, HPV18, and Ki67 combination showed the excellent combination to study as the early expressions in the form of high-concentrations followed by MCM5 and CD146. Figure 2 demonstrates the basic steps for sample processing.

The cervical cancer all biomarkers, were done for the protein quantification using BCA method and through an ELISA method in the prospectively collected cervix swab samples as per the inclusion and exclusion criteria of the study.

Sandwich ELISA experiments

Sandwich ELISA performed to detect p16NK4A, Ki67, HPV16, HPV18, MCM5 and CD146 protein as biomarkers for cervical cancer detection in the tested population and control as well. A human p16NK4A, Ki67, HPV16, HPV18, MCM5 and CD146 protein ELISA Kit (Biocompare) was used here with recommended protocol. Here, 96 ELISA plates already coated with 100 µl anti-respective biomarkers antibodies (20 µg/ml) were incubated with fivefold (500 µl) cervical swab cell lysate. The cell lysate washed two times with 200μL Phosphate-buffered saline + 0.05% Tween20 (PBST). Detecting antibody incubated for one hour at 370C. The secondary antibody (1:5000 dilutions in PBS) was added to each plate 100 µl, 20 µg/ml, and incubation was allowed for one hour at 37 °C. The plates were washed three times with 200 µl PBST and add 3, 3′, 5, 5′-Tetramethylbenzidine solution to each well and incubated for 30 min. Reaction was stopped using 100 µl 1 M sulphuric acids, and absorbance was measured at 450 nm using an ELISA reader.

Diagnostic device KayaCervi

The Prototype A (Vega 200) named KayaCervi, developed at the National Tertiary Care Hospital of central India with the aim of mass screening for cervical cancer was validated as per the Grant-in-Aid-DHR grant (R.11016/02/2023-GIA/HR).

The Kayacervi system consists of a Reader (Vega200), KayaCervi Test kit and comes with a computer with Vega Software. Primary kits are assembled and tested with real samples. The foot print is 6″ × 12″ × 3″. The cartridge is made of polystyrene, with a custom etched membrane, that is conditioned to bind the antibody and protein complex on the membrane (Figs. 3, 4 and supplementary S1 and S2).

Fig. 3

Diagnostic device KayaCervi

Fig. 4

Diagrammatic representation of the device functioning based on Antigen – Fluorescently-Tagged Antibody reaction

The Test kit contains reagents comprising of fluorescent probe, Extraction Buffer, Wash buffer and read Buffer. The principle of the test is based on selective filtration. The unfiltered sample probe complex binds to the cartridge and is imaged with the reader.

Technology utilized in Kayacervi (under Kaya17) is robust and has been validated successfully for testing various proteins, viruses and food-borne pathogens (U.S. Patent Application Serial No. 16/690,589 and 16/690,589; Indian patent application number 201941011177 and 201841044093). Amplification of the signal in this technology based on immunodiagnostic principle is not by one source but a combination of the type of fluorophore, the cartridge and the reader. Kaya17 is a Rapid Point of Care reader, which works off the USB port of a computer. It does not need to be plugged into a power supply.

The KayaCervi device (Supplementary S1) is as a "high-throughput, multi-wavelength fluorometer optimized for real-time immunoassay tracking. The advantage of this is that Excitation and Emission matrix (EEM) can be used instead of standard single-wavelength fluorescence. Conventional readers miss shifts in the microenvironment but an EEM based captures the entire spectral landscape, allowing seeing changes in both intensity and wavelength simultaneously. Especially an EEM based POC diagnostic device focuses on portability, speed, and automation, because it operates at the patient's bedside or in the field.

Working Principle is:

$$\begin{aligned} & \left[ {{\text{Target Antigen}}} \right]{\mkern 1mu} + {\mkern 1mu} \left[ {{\text{Fluorescently}} - {\text{Tagged Antibody}}} \right] \\ & \quad \quad \quad \quad \quad \quad \quad \downarrow \\ & \quad \quad \quad \quad \left[ {{\text{Binding }}/{\text{ Reaction}}} \right] \\ & \quad \quad \quad \quad \quad \quad \quad \downarrow \\ & \quad \quad \left[ {{\text{Micro environmental Change around Fluorophore}}} \right] \\ & \quad \quad \quad \quad \quad \quad \quad \downarrow \\ & \quad \quad \left[ {{\text{Shift in EEM Landscape }}\left( {{\text{Ex}}/{\text{Em Peaks}}} \right)} \right] \\ \end{aligned}$$

  • The Immune Reaction: The target antigen binds to the fluorescence-tagged antibody.

  • The Spectral Shift: Binding alters the local chemical environment (polarity, rigidity, or proximity quenching) of the fluorophore tag. This causes a distinct shift or intensity change in its excitation-emission peak.

  • Data Generation: The device rapidly scans the sample, plotting a 3D EEM contour map to quantify the exact ratio of bound to unbound antibodies.

  • Light Source: High-speed tunable source a fast-switching LED array to sweep through excitation wavelengths quickly.

  • Detection System: A spectrograph paired with a Charge-Coupled Device (CCD) camera to capture the entire emission spectrum instantly for each excitation step.

  • Microfluidics/Sample Handling: A microfluidic chip channel to handle low-volume Ab-Ag samples efficiently.

Our primary hypothesis is that sensitivity and specificity of Kaya17 for cervical cancer screening is equal to that Pap smear or VIA. We are considering the confidence limit of 95% and power of 80% for detection of any difference in diagnostic accuracy measures (sensitivity, specificity etc.). We are using a multistage cluster sampling technique to select study participants from the villages in Bhopal district. Using the anticipated parameters, the sample size for sensitivity is 865 and for specificity are 45.

Liquid based cytology (LBC)

Relevant clinical details and patient history were obtained from standardized requisition forms accompanying each sample. Cervical specimens were collected exclusively by trained gynaecologists. Prior to sample collection, all participating clinicians received instruction on the manufacturer’s guidelines for liquid-based cytology (LBC) to ensure uniformity of technique. Patient information sheets in the local language and further consent form taken prior to their swab sample collection (IHEC – LOP/2022/EL071).

Sample collection and processing

Liquid-based cytology samples were collected using the EZIPREP® LBC system (LBC India). After cervical sampling, the brush head was detached and immersed directly into EZIPREP® vials containing a buffered alcohol-based preservative solution. Samples were then transported to the pathology laboratory for processing.

In the laboratory, LBC specimens were processed using the EziPrep® LBC Nanocyte Mini device. This system employs a sedimentation-based enrichment method in which the samples are centrifuged to obtain a cell pellet. A portion of the enriched pellet is subsequently transferred into an Eziprep® Premium Opti-Coated slide—a pre-coated microscopy slide designed specifically for EziPrep LBC preparations. The device produces a uniform thin-layer monolayer preparation approximately 13 mm in diameter. All prepared slides were stained using the Papanicolaou (Pap) staining technique (Image enclosed).

Cytological evaluation

All slides were examined by experienced cytopathologists. The cytological findings were categorized and reported according to The Bethesda System (TBS) 2014 for reporting cervical cytology.

Immuno-histochemistry (IHC)

IHC carried out to validate the expression of biomarker proteins expression in tissue samples whos KayaCervi and Pap smear tests were positive. For IHC patients, biopsy samples used. The tissue section was sliced i.e., 5 cm, followed by de-paraffinization and fixed with formalin solution and embedded with paraffin. Further, the tissue sections were stain with 200 µl, 18 µg/ml ki67 and p16 specific antibodies. Furthermore, tissues were incubated with a biotin-conjugated secondary antibody (1; 1000), followed by 200 µl, 20 µg/ml horseradish peroxidase (HRP-conjugated streptavidin (Dianova). The tissue sections were counterstained with hematoxylin to visualize the nuclei. Similar protocol was used for P16, Ki67, HPVs HIC. (Supplementary S4: IHC result and biomarker based results of the carcinoma cervix of a patient).

IHC was subsequently performed for cases identified as positive by LBC to provide additional pathological confirmation of positive findings as per our institutional protocols. IHC was therefore used as a confirmatory investigation of positive cases, rather than as the universal reference standard for all participants.

Results

Phase I (for Standardization at the hospital setting) for the device validations using different biomarkers and tested for cervical cancer by the sandwich ELISA test and on KayaCervi device method and simultaneous compared results with LBC, the gold standard Pap-Smear and IHC for establishing its sensitivity and specificity at AII India Institute of Medical Sciences Bhopal, India. Both test and control swab samples were processed, and cell lysate was prepared (Supplementary S3). The analytical evalution and diagnostic performance parameters, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), overall accuracy, and agreement analysis.

  • Phase I: Reference-based analysis and assay standardization, including clarification of the sample composition and comparison of KayaCervi results with the reference method.

  • Phase II: Community-based evaluation of KayaCervi and detailed diagnostic performance analysis.

  • Pap smear analysis: Clearly described as the reference cytological method used for the relevant Phase I comparisons.

  • LBC analysis: Clearly described as the cytological reference method used during the Phase II community-based evaluation.

  • IHC analysis: Clearly described separately as confirmatory analysis performed for selected positive cases and not as the universal reference standard for the entire study population.

Further, Phase II (Community Validation) total 867 (66 samples degraded to do analysis of n = 801) cervical swab samples collected from the > 30 years female apparently healthy from the community for the mass screening using KayaCervi Device and simultaneously compared the results with LBC for the specificity and sensitivity, out of which 66 swab samples were could not be processed due to degradation of the samples. Real-world evaluation in 801 apparently healthy women undergoing simultaneous LBC testing, with particular emphasis on specificity and feasibility for population-based screening.

The prevalence of cytological abnormalities in the screened population (1.16%) was lower than the 6.1% symptomatic women prevalence used for sample size estimation. Consequently, specificity estimates were robust, whereas sensitivity estimates should be interpreted with caution and warrant further evaluation in larger cohorts with a greater number of positive cases.

Instrument and validation

Prototype A (Quantum 1001) KayaCervi of the instrument is developed and under a rigorous clinical setting. Study has tested a suspected 200 cancer samples from a cervical swab in the test population at the hospital setting (Table 2, Fig. 5) and conducted community based study.

Table 2 Total Number of Cervix Swab samples tested (Hospital setting for the standardizing the test).

Full size table

Fig. 5

Correlates well with ELISA test as shown with R2 Values of 0.87

The novelty lies in the technology is in design. It doesn’t require any lab infrastructure, Device is made up of Cartridge, Reader, Chip and reagents, Based on well known-Fluorescence and antigen–Antibody reaction. Cartridge is designed to have selective filtration that help capture the entire sample during imaging, has vertical flow and enables quantification of the sample. Cartridge contain fluorescence tagged antibody, bounded antigen–antibody complex on cartridge, inserted in reader.

Excited and emission captured to convert to results with the software, measured as Relative fluorescence units (RFU) (excitation-emission matrices (EEMs) (Principles of fluorescence spectroscopy by Lakowiez18).

In Phase I and Phase II study pathway followed as: Candidate biomarker evaluation by ELISA → comparison with Pap smear → biomarker selection → KayaCervi assay development/optimization → threshold establishment → community-based field evaluation.

Phase I cohort constituted an independent clinical validation dataset. It represents assay development/standardization.

Phase I: Standardization phase

  • 200 screening participants 5 samples were LBC and KayaCervi Positive.

  • In these 200 sample 20* confirmed positive cases included to optimize assay performance and establish cut-off values.

  • Sensitivity explored during device development.

  • Total samples (N): 200

  • Observed agreement (Po): 100%

  • Expected agreement (Pe): 78.13%

  • Cohen's Kappa (κ): 1.000

Analysis indicated A κ = 1.000 indicates perfect agreement between the device and the reference standard with.

  • Sensitivity: 25/(25 + 0) = 100%25/(25 + 0) = 100%25/(25 + 0) = 100%

  • Specificity: 175/(175 + 0) = 100%175/(175 + 0) = 100%175/(175 + 0) = 100%

  • Positive Predictive Value (PPV): 100%

  • Negative Predictive Value (NPV): 100%

  • Overall Accuracy: 100%

Above diagnostic performance analysis is showing A Cohen's kappa coefficient of 1.00, indicating complete concordance. The device demonstrated 100% sensitivity, 100% specificity, and 100% overall accuracy in the validation panel of 200 known samples. Agreement between the device and the reference method was perfect. The ELISA and KayaCervi results were evaluated against the Pap smear classification. These analyses were intended to establish the relationship between the candidate biomarkers, the selected biomarker combination and the cytological classification, and to support development of the point-of-care assay. They were not intended to constitute an independent clinical validation of the final device. In the Phase I assay-development/standardization cohort of 200 samples, including 5 reference-positive and 195 reference-negative samples, KayaCervi showed an observed sensitivity of 100%, specificity of 92.31%, and overall accuracy of 92.5% relative to the Pap smear reference classification. As this cohort was used during assay optimization and threshold establishment, these estimates should not be interpreted as independent clinical validation. R2 analysis of each biomarkers is added in the annexure section (Table 3).

Table 3 Showing Phase I standardization data analysis.

Full size table

ELISA based analysis

Here, Fig. 5 shows linear progression between RFU value in the KayaCervi system and ELISA test for Biomarkers tested samples with an R2 value of 0.87 (Figs. 6, 7, 8).

Fig. 6

ELISA-based protein concentrations of Ki67, p16INK4a, HPV16 and HPV18 in cervical swab samples from women classified as high-, low- or no-risk for cervical cancer. Box-and-whisker plots depict median, interquartile range and overall range. Ki67, HPV16 and HPV18 levels are significantly higher in the high-risk group (*P < 0.05, ***P < 0.001), while p16INK4a shows no significant difference (ns).

Fig. 7

ELISA-based protein concentrations of candidate biomarkers in cervical swab samples used for cervical cancer screening. Box-and-whisker plots show the distribution of Ki67, p16INK4a, MCM5, HPV-16, HPV-18, CA125, CA19-9, ACTN4 and CD146 levels (ng/ml), with boxes representing the interquartile range, horizontal lines the median, and whiskers the minimum and maximum values.

Fig. 8

KayaCervi point-of-care measurements of Ki67, p16INK4a, HPV16 and HPV18 in cervical swab samples, with women stratified into high-, low- and no-risk groups according to device readings. Box-and-whisker plots show higher Ki67, HPV16 and HPV18 levels in the high-risk group, while p16INK4a is increased in both high- and low-risk groups compared with the no-risk group, with no difference between high and low risk (ns). Group differences were tested using one-way ANOVA followed by Tukey’s post hoc test, with significance set at P < 0.05.

ELISA based protein concentration

Sensitivity, Specificity, Positive Prediction Value (PPV) and Negative Prediction Value (NPV) to compare the tested biomarkers (Supplementary S3).

Immuno-histochemistry (IHC) and LBC

Further, immune-histochemistry (IHC) was performed for the positive cases. As a result, shown in (supplementary S4) these tissues possess a higher expression Ki67, P16 over control. LBC testing of a patient suspected with cervical cancer are also showing in the Figs. 9, 10, 11.

Fig. 9

Showing HSIL (High-Grade Squamous Intraepithelial Lesion) in 10 × objective lens revealed the presence of abnormal, large nuclei are seen that are considerably bigger than the intermediate cell nuclei, hyperchromatic group and dysplastic cells with metaplastic cytoplasm and irregular nuclear contours.

Fig. 10

Liquid based cytology testing of a patient suspected with cervical cancer showing High grade squamous cell intraepithelial lesions (HSIL) is also detected through KayaCervi device

Fig. 11

Sankey Diagram (Conceptualization) showing distribution of KayaCervi risk groups in relation to the LBC.

As per the above results and values set as the cut-off values to interpret the results, we conclude that biomarkers p16iNKA4, Ki67, HPV-18, and HPV16; can be considered predictive biomarkers. The presence of HPV 16 and 18 with or without Ki67 and P16iNKA4 presence or not, would indicate the High risk (HR) group and such woman can be suggested of taking up the further medical advice to take precautions. If a sample positive for Ki67 and P16iNKA4 would come under moderate risk (MR) group which is comparable to the LBC’s RCC findings. All four biomarkers absence will indicate no risk (NR) group. The early non-cancerous stage of cervical cancer may lead to the cancerous stage as mild to severe CIN but if not monitored or screened for these early signs of the disease will be crucial for resource limited rural arears.

For further analysis we have considered KayaCervi HR reported (26) with LBC positive 10 (ASCUS n = 8, ASC-H n = 01 and LSIL n = 01) samples collected from the community (n = 801) (Supplementary S6 Table 1, S7 Table 2, S8 Table 4).

Table 4 Antibody specificity and sensitivity of the selected 10 antibodies.

Full size table

  • Total samples (N): 801

  • Observed agreement (Po): 98.0%

  • Expected agreement (Pe): 95.6%

  • Cohen's Kappa (κ): 0.545

Thus, κ = 0.545 indicates moderate agreement between the device and the reference standard (Assistance taken from the AI tool for data calculation). The Diagnostic performance calculated from the generated data is :

  • Sensitivity: 10/(10 + 0) = 100%10/(10 + 0) = 100%10/(10 + 0) = 100%

  • Specificity: 775/(775 + 16) = 97.98%≈98.0%775/(775 + 16) = 97.98\% \approx 98.0%775/(775 + 16) = 97.98%≈98.0%

  • Positive Predictive Value (PPV): 10/(10 + 16) = 38.5%10/(10 + 16) = 38.5%10/(10 + 16) = 38.5%

  • Negative Predictive Value (NPV): 775/(775 + 0) = 100%775/(775 + 0) = 100%775/(775 + 0) = 100%

  • Accuracy: 785/801 = 98.0%785/801 = 98.0\%785/801 = 98.0%

During analytical validation using 200 known samples, KayaCervi demonstrated complete agreement with the reference method, yielding 100% sensitivity, 100% specificity, and a Cohen's kappa coefficient of 1.00, indicating perfect agreement (Table 5).

Table 5 Showing Phase II Validation data analysis.

Full size table

In community-based screening of 801 women, KayaCervi identified all 10 reference-positive cases and correctly classified 775 of 791 reference-negative cases. The device demonstrated a sensitivity of 100%, specificity of 98.0%, positive predictive value of 38.5%, negative predictive value of 100%, and an overall diagnostic accuracy of 98.0%. Cohen's kappa coefficient was 0.545, indicating moderate agreement beyond chance (Tables 6 and 7). In the community cohort, estimate is based on only 10 positive cases in apparently healthy females and therefore has limited precision but this has detected during the screening shows the Kayadevice efficiency.

Table 6 Agreement analysis.

Full size table

Table 7 Confusion matrix (Community screening).

Full size table

In community-based evaluation (N = 801), the device demonstrated an overall agreement of 98.0% with the reference standard. Cohen's kappa coefficient was 0.545, indicating moderate agreement beyond chance. The device achieved 100% sensitivity, 98.0% specificity, 38.5% positive predictive value, 100% negative predictive value, and 98.0% overall accuracy. However, it would be appropriate to note that while specificity was estimated with good precision, agreement and sensitivity should be interpreted cautiously because of the small number of positive cases found during the community mass screening in the rural areas in the study. However, in LBC analysis, RCC percentage (Bacterial or Fungal infections) (45.69%) in the healthy women were high in percentage, can be comparable to the MR group of KayaCervi device identification in the premenopausal women.

In Fig. 12 Showing ASC-H, (Atypical squamous epithelial cell—High Grade squamous intraepithelial lesion revealed the presence of small immature squamous metaplastic cells with dark but smudgy nuclear chromatin under 10 × objective lens.

Fig. 12

Apparently healthy females mass screening (n = 800) showing analysis of KayaCervi Device parallel compared with the LBC test. KayaCervi Device HR (High Risk when HPV 16 and 18 are Positive), KayaCervi MR (Moderate Risk, when HPV Negative but Ki67 and p16 Positive) KayaCervi NR No Risk (NR) when all four biomarkers are Negative). For LBC based testing (LSIL: Low-grade squamous intra-epithelial lesion, HSIL: High-grade squamous intra-epithelial lesion, NILM: Negative for intra-epithelial lesion or malignancy, RCC: Reactive cellular changes, ASC-US: Atypical squamous cells of undetermined significance, ASC-H: Atypical squamous cells cannot exclude HSIL. (Data is Generative AI tools based). Sankey Diagram (Conceptual illustration)

Discussion

Screening for cancer is known to reduce mortality by early detection and treatment19,20,21. However, there are two prerequisites for screening to reduce the rate of death from cancer. First, screening must advance the time of diagnosis of cancers that are destined to cause death. Second, early treatment of these cancers must confer some advantage over treatment at clinical presentation. Unlike other cancer sites, cervix can be subjected to screening for early diagnosis and treatment. However, despite availability of gold standard method Pap-smear for cervical cancer diagnosis present at mainly division level there is no countrywide government-sponsored public health policy on prevention of cervical cancer by either mass screening or vaccination or both (ICMR 2016; WHO 2002). Carcinoma of cervix is a major health problem faced by Indian Women. The incidence of cervical cancer is higher than endometrial and ovarian cancer (Mathew and George 2009)22,23,24. There were an estimated 528,000 cases and 266,000 deaths from cervical cancer worldwide in year 2012 which accounts for 7.5% of all female death from cervical cancer25,26,27. India accounts for about 20% of cervical cancer cases reported worldwide. Persistent infection by high-risk human papillomavirus (HPV) accounts for nearly 95% of cervical cancer cases. Specifically, HPV16 and 18 collectively contribute to about 70% of cases28. In that view of the existing challenge is to develop diagnostic device for mass-screening, which should be low in cost and ideal for rural and primary health centers of India for early detection8,29,30,31,31,32,33,34,35,36,37,38,39,40 The development of a successful medical device requires not only engineering design efforts, but also clinical, regulatory, marketing and business expertise. In coming decade, medical technology innovations will fundamentally transform the health care landscape, providing new solution to address diseases like cervical cancer and revolutionize the way treatments are administered. India being a developing country with different demographic and socio-economically to other developed countries, low-cost cervical cancer diagnostic kit for mass screening would encourage suspected women to get screened for cervical cancer would be a step towards an early detection and further treatment is the need of an hour41,42,43. Current recommendations of the American Society for Colposcopy and Cervical Pathology (ASCCP) state that women ages 30 years and older who have normal cytology but are high risk HPV DNA positive may benefit from genotyping assays for the presence of HPV 16 and 1844,45,46,47. Women in whom HPV 16 and 18 is detected should be referred for colposcopy. HPV DNA testing provides a quantitative means of HPV detection, compared to the valuation of cellular changes in cervical cytology, which is more subjective. HPV DNA testing is also performed as a reflex test on any ASCUS pap smear. HPV testing has saved both patients and the health care system a significant amount of time and resources, but the system remains cumbersome, requiring multiple visits6,48,49,50,51

Pap tests, HPV tests, and VIA are used for screening in healthcare facilities. Since 2017, VIA has been used for population-based or mass screening as part of the National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Disease, and Stroke (NPCDCS)7,52. Late diagnosis, insufficient screening coverage, limited access to Pap Smear labs, expense of test, slow turnaround time, several hospital visits, skip routine screening, For Pap tests, there are challenges with handling and transporting slides, keeping cell samples safe, and waiting for results before clinicians can examine them8. Having trouble bringing ladies back for treatment and confirmatory diagnostic testing the high incidence of cervical cancer in women in India is primarily due to problems with VIA, which are subject to prejudice and subjective interpretation. Almost all cervical cancer cases (99.7%) are linked to infection with high-risk HPV, an extremely common virus transmitted through sexual contact53, WHO). CIN is a premalignant lesion that may exist at any one of three stages: CIN1, CIN2, or CIN3. If left untreated, CIN2 or CIN3 can progress to cervical cancer54,55,56,57,58. Successful cancer treatment mainly relies on early and effective diagnosis59,60.

In that view of the existing challenge is to develop diagnostic device for mass-screening, which should be low in cost and ideal for rural and primary health centers of India for early detection8. Information regarding the cervical cancer biomarkers is available mainly for the research purpose and their validations to develop the screening device have not been explored yet. The low-income nations with in-adequate diagnostic and medical facilities are associated with a higher prevalence of different types of cancer61.

The level of p16NK4A, Ki67, HPV16 and HPV18 was determined via a sandwich ELISA, LBC and Pap smear tests reconfirmed by IHC. Squamous metaplasia is a preferential site for high-risk HPV infection and HPV related carcinogenesis and thus is a hot spot for developing squamous intraepithelial lesion (SIL) and squamous cell carcinoma62. However, it is a physiologic condition and is not a precursor per se. Colposcopy can identify squamous metaplasia as transformation zone, which is a site for targeted biopsy in case of abnormal Pap smear63. Transitional cell metaplasia is a mimic of squamous metaplasia and high grade SIL (HSIL) and is a form of atrophy. It should be kept in mind that squamous metaplasia, in particular its immature form, can be erroneously misinterpreted as HSIL. The p16 immunohistochemistry, a diagnostic tool for HPV related neoplastic condition including HSIL, can solve the problem since squamous metaplasia is negative for this particular maker64,65

This study aims to translate these molecular findings into a diagnostic test kit for cervical cancer. Examining the HPV16, HPV18, p16NK4A, Ki67, MCM5, CD146, CEA, CA125, CA19-9 and ACTN-4 expressions in cervical swab samples individually and out of these four HPV16, HPV18, p16NK4A, Ki67 biomarkers in combination were used as the predictive markers for the mass screening device to cervical neoplasms. Development of KayaCervi device is community-based mass screening using multiple biomarkers of cervical cancer emphasizing specificity is scientifically reasonable because, the vast majority of screened women are disease-negative, False positives lead to unnecessary anxiety, referrals, repeat testing, colposcopy, and increased healthcare costs, a screening device deployed at the population level must demonstrate a high ability to correctly identify healthy women44,58,66,67,68

Study was conducted in the central India state Madhya Pradesh Bhopal city and its surrounding villages. In Phase I device validation with a Cohen's kappa coefficient of 1.00, indicating complete concordance. The device demonstrated 100% sensitivity, 100% specificity, and 100% overall accuracy in the validation panel of 200 known samples. Although in real community testing (N = 801), the device demonstrated an overall agreement of 98.0% with the reference standard. Cohen's kappa coefficient was 0.545, indicating moderate agreement beyond chance. The device achieved 100% sensitivity, 98.0% specificity, 38.5% positive predictive value, 100% negative predictive value, and 98.0% overall accuracy. However, while specificity was estimated with good precision, agreement and sensitivity should be interpreted cautiously because of the small number of positive cases. The Phase I cohort consisted of 200 samples, of which 5 were reference-positive by Pap smear and 195 were reference-negative. Thus, the 20 samples classified as positive by KayaCervi should not be described as 20 confirmed positive cases. The diagnostic estimates should instead be interpreted based on the 5 reference-positive and 195 reference-negative participants. The ELISA and KayaCervi results were evaluated against the Pap smear classification (WHO 2012). These analyses were intended to establish the relationship between the candidate biomarkers, the selected biomarker combination and the cytological classification to support development of the point-of-care assay. They were not intended to constitute an independent clinical validation of the final device.

The primary objective of the community-based validation study was to evaluate the performance of the device as a mass screening tool among apparently healthy women. In such settings, the majority of participants are disease-negative, and high specificity is essential to minimize false-positive results, unnecessary referrals, psychological distress, and healthcare costs. Therefore, the validation study focused primarily on the assessment of specificity in a real-world screening population. Sensitivity assessment was undertaken during the device standardization phase using both screening participants and additional confirmed positive cases to ensure adequate representation of disease-positive samples for assay optimization.

Validation of the KayaCervi prototype using suspected 200 cervical swab samples collected swab samples at the hospital setting in the phase I, demonstrated comparable efficacy to Pap smear tests. The prototype offers advantages for mass screening in developing countries like India, where the Pap smear test may not be available at the rural healthcare units. The p16NK4A, Ki67, HPV 16 and HPV18 combination-based test devices present an affordable and efficient solution for widespread cervical cancer screening. The device's portability enables on-field testing; a significant advantage over hospital-based Pap smear tests69. The global POC diagnostics market, valued at USD 23.16 Billion in 2016, is projected to reach USD 36.96 Billion by 2021, with a CAGR of 9.8%. The diagnostic market for cancer, including cervical cancer, reports higher growth in the Asia–Pacific region70,71,72,73.

For Phase II, the sample size was calculated using a previously reported prevalence of 6.1% for cervical cytological abnormalities. However, the observed prevalence of abnormal LBC findings in the screened population was lower (1.16%; 10/865). Consequently, specificity estimates were derived from a large number of disease-negative participants and are therefore expected to be robust. In contrast, sensitivity estimates were based on a relatively small number of positive cases and should be interpreted with caution, as low disease prevalence is known to limit the precision of sensitivity estimates in diagnostic accuracy studies74.

There are several well-recognized methodological references that support the statement that when the number of disease-positive cases is small, sensitivity estimates become imprecise, whereas specificity can be estimated more reliably when there are many disease-negative participants75. Nevertheless, the PPV was only 38% reflecting the low prevalence of reference-positive cases in the screened population and the presence of 16-false-positive results. Therefore, suggesting PPV as an intrinsic property of the device is dependent on the disease prevalence. IHC was conducted only in LBC-positive cases, as they cannot be used to calculate an unbiased sensitivity and specificity for the entire Phase II cohort. The IHC’s role here is a confirmatory testing for the LBC positive cases not as a reference standard for the overall diagnostic accuracy. Previously done study has reported that LBC and HPV screening is effective in identifying high grade lesions (Zhoa et al. 2022)76,77,78,79,80.

The innovation in KayaCervi lies in its lab-free design, USB supported, comprising a Cartridge, Reader, Chip, and reagents utilizing Fluorescence and antigen–antibody reactions. The Cartridge, featuring selective filtration for complete sample capture, enables quantification with vertical flow. Fluorescence-tagged antibodies in the Cartridge form the antigen–antibody complex, read by the Reader. The technology uses Excitation-Emission Matrices (EEMs) for Relative Fluorescence Unit (RFU) measurement (Supplementary S1)81,82. The Reader scans a wide sample area, combining signals into the system. KayaCervi, consisting of the Vega-200 Reader, PC with Vega Software, and Test kit, allows mass screening in basic healthcare settings. Phase I and Phase II validations in the study showing specificity 90% and sensitivity comparable to Pap smear. Presently available diagnostics for a high-grade lesion, the sensitivity of a single Pap smear is 60–75% are inherent in cytology with false-positive rates of 15–50% and false-negative rates of 30% needs infrastructure. The overall sensitivity ranged from 30 to 99% and the specificity ranged from 86 to 100%. In is reported in that for LSIL/CIN-I, the overall specificity ranged from 86 to 100% and the sensitivity ranged from 30 to 87%; in 8 studies sensitivity was less than 60%. For LSIL/CIN-II-III, the overall specificity ranged from 91 to 98% and the sensitivity ranged from 44 to 99%83, indicating that KayaCervi device would be good for the mass-screening of the community for the cervical cancer84

In our project, we evaluated 10 antibodies initially in the cervix swab samples, including CD-146, MCM5, CA19-9, CEA, ACTN4, P16INK4A, HPV16, HPV18, CA 125, and KI67, of which HPV-16, HPV-18, KI67, and P16INK4A demonstrated high sensitivity and specificity. A community-based study is under taken for the mass screening of cervical cancer device KayaCervi validations. Villages near Bhopal city (central India) and Seoni district have been covered. As per the study 36 villages are selected using probability Proportionate to size sampling (PPS). 65% Sensitivity, 70%specificiity, 6.1% prevalence (Symptomatic females), 95%CI, 20%Relative precision, design effect of 1.5 and 10% included as non-response accordingly 865 samples screened to fully validate the device. However, the sample size was calculated using a previously reported prevalence of 6.1% for premalignant and malignant cervical lesions in the target population. Based on this estimate, a minimum sample size of 865 participants was required for diagnostic accuracy evaluation. However, the observed prevalence of cytological abnormalities in the study population in apparently healthy females was lower, with 10 participants (1.16%) demonstrating abnormal LBC findings (ASCUS, ASC-H, or LSIL) that is alarming as detected by the KayaCervi device for the timely intervention. It was also observed that females living in urban has incidences because they have significantly better access to health education and facility (Yesilfidan et al. 2025; Khanna et al. 202485,86,87,88,89) thus it makes very crucial to mass screen rural apparently healthy women. KayaCervi device analysis showing MR group is mostly indicating the LBC test results as RCC outcomes indicating towards the infections (45.69%)90. High infections in the rural population is one of the concerning factors, as these infections either bacterial or fungal if remain untreated can cause cytological changes in the cervix or compromise individuals immunity to get HPV infections in view of low levels of hygiene (Godoy-Vitorino et al. 2018). It is reported that fungal infections may increase the risk of HPV infection and promote virus persistence (Pietro et al. 2018)70. Women are more susceptible to infection with the HPV, and the most infected are girls in puberty and women under 20 years of age. In elderly women over 55 years of age, this may be related to the persistence or reactivation of a previously acquired infection, rather than new or recent infections71,84. Consequently, specificity estimates were derived from a large number of disease-negative participants, whereas sensitivity estimates were based on a smaller number of positive cases and should therefore be interpreted with caution. We also faced the challenges in taking cervix swab samples from the rural areas’ women. Limited patient compliance poses a significant challenge in cervical cancer screening programs, as some patients may be reluctant to follow recommended sample collection procedures. This reluctance can lead to low patient participation in spite of multiple rounds of camps and a failure to adhere to follow-up testing, ultimately reducing the accuracy of screening efforts. To address this issue, it is crucial to improve their comfort during the collection process, and consider implementing non-invasive methods, such as developing self-swabbing sticks or cervix swab less screening using urine sample may help.

Limitations

The present work represents a substantial translational effort undertaken under real-world community-based screening conditions. The study involved the development of a multiplex biomarker-based point-of-care platform, assay standardization, biomarker evaluation, and subsequent field evaluation in a large community cohort. Conducting cervical cancer screening in community settings, particularly in populations with limited access to conventional laboratory and specialist diagnostic facilities, involves considerable logistical, technical and operational challenges. The successful implementation of the KayaCervi platform in this setting therefore provides important preliminary feasibility and field-performance data. Findings were obtained from the present study population and that independent validation in geographically and clinically diverse populations is required before definitive conclusions regarding the generalizability and clinical performance of the device can be made. One of the limitations is that its independent external validation is essential; the present study provides an important foundation for further clinical translation of the device. The findings support the potential utility of KayaCervi as a portable, biomarker-based cervical cancer risk-screening platform that may be particularly relevant for primary health-care settings and low-resource or decentralized health-care facilities, where access to conventional laboratory-based screening and specialist cytological services may be limited.

Conclusions and future scope

This study validated KayaCervi, a portable, laboratory-independent point-of-care diagnostic device developed to address India's high burden of cervical cancer. The device integrates a fluorescence-based detection platform with a four-target biomarker panel comprising p16INK4a, Ki-67, HPV16, and HPV18, achieving a specificity of 86–90% in community-based rural screening. As a rapid, low-cost, and user-friendly screening tool, KayaCervi has the potential to democratize access to cervical cancer screening by overcoming barriers related to laboratory infrastructure, trained personnel, and patient follow-up. Further, it include the need for larger, prospective, multicenter studies involving geographically diverse populations, with the biomarker panel and diagnostic thresholds predefined before evaluation and with an appropriate independently established reference standard. Such studies will be essential to confirm the reproducibility, generalizability, diagnostic performance, clinical utility and implementation feasibility of KayaCervi before large-scale deployment.Cervical cancer is one of the major contributors to oncological mortality among Indian women in the Indian public health scenario with a regional prevalence alarmingly high (6% to 30%). This heavy disease burden is intrinsically linked to the foundational issues of community medicine, particularly in rural and underserved areas where health literacy is low, socioeconomic barriers are high, and systemic inequities in healthcare continue. The community health perspective faces huge operational bottlenecks in the implementation of effective mass screening. The existing health infrastructure is highly centralized and PHCs and rural sub-centres are ill-equipped to handle traditional diagnostic modalities. LBC and standard Pap smears are reliable diagnostic pillars. But at grass root level, these are virtually non-existing due to high cost, specialized laboratory dependencies and need for cytopathologists. To overcome this, the government-sponsored National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke (NPCDCS) has depended on Visual Inspection with Acetic Acid (VIA) for population-based screening. However, practitioners of community medicine often face the shortcomings of VIA, which is plagued by subjective interpretation, high inter-observer variability and the tendency to produce false results—all of which diminish its potential to reduce long-term mortality.

The study successfully developed KayaCervi, a novel lab-free point-of-care diagnostic device for large-scale on-site population screening, specifically designed to fill this critical infrastructure gap. The portable platform, powered simply via a USB port, integrates a specialized cartridge, reader and microchip system. It provides an alternative to the requirement for formal laboratory infrastructure by employing fluorescence-based antigen–antibody detection using Excitation-Emission Matrix (EEM) technology. Although the device initially tested a broad panel of ten possible biomarkers for cervical cancer, rigorous diagnostic optimization narrowed the final array down to a highly effective four-biomarker combination: p16INK4a, Ki67, HPV16 and HPV18. This particular multi-marker approach provides a multi-dimensional diagnostic perspective. The tool tracks p16INK4a to detect the primary viral drivers, HPV16 and HPV18, captures active oncogenic viral transformation and measures abnormal cellular proliferation via Ki67.

The KayaCervi platform was rigorously validated in a two-phase design to demonstrate its efficacy in clinical and real-world settings. Phase I was performed for standardization and initial validation with swab samples from a cohort of 200 symptomatic/suspected patients in a tertiary care hospital setting with direct benchmarking of the device with ELISA, Pap smears and liquid-based cytology. Phase II brought the technology into its target environment: mass screening in a community setting. The study followed a Probability Proportionate to Size (PPS) model to screen a healthy and symptom free cohort of 801 women from 36 different villages of Bhopal and Seoni districts of Madhya Pradesh. High prevalence of infections was noted in the premenopausal apparently healthy females, defining the crucial role of the mass screening at the community level.

The community validation was intentionally skewed toward very high specificity, consistently between 86 and 90%, from a statistical and public health perspective. Specificity is important in mass-screening settings where the majority of the population is asymptomatic and disease-negative. By reducing false positives, vulnerable rural populations are directly protected from unnecessary psychological distress, low-income families are saved from debilitating out-of-pocket diagnostic costs and the over-saturation of tertiary healthcare referral systems is prevented. Interestingly, the study sample size was based on an expected historical premalignant prevalence of 6.1%, but the actual observed prevalence of cytological abnormalities in the field was observed low at 1.16% (10 out of 867 participants) in the studied apparently women and detected. This low-prevalence scenario enabled a highly reliable, robust estimate of the specificity of the device. On the other hand, it made the sensitivity metrics (from 40 to 80% depending on the specific biomarker combinations) more mathematically imprecise—a common diagnostic variance well documented in low-prevalence epidemiological literature.

Although, the device study did show definite implementation challenges with patient adherence in rural settings as healthy participants were often initially reluctant due to modesty, cultural taboos and physical discomfort surrounding the collection of pelvic swabs, which required extensive community camp outreach to overcome. This operational feedback suggests that to optimize future field deployment, non-invasive collection alternatives need to be developed such as self-swabbing sticks or engineering the platform to perform cervix-swab-less molecular screening using urine samples. Ultimately, with the due limitations; KayaCervi is a fast, affordable, highly specific, field deployable solution for translating complex molecular biomarker findings into an actionable public health tool. This technology removes the structural limitations of traditional pathology labs and offers a practical roadmap to democratize early cancer detection across rural India, directly supporting community medicine initiatives to make cervical cancer a preventable disease.

References

  1. Yang, S. T. et al. Cervical cancer: Part I human papilloma virus vaccination in Taiwan. Taiwan J. Obstet. Gynecol. 63(3), 320–328. https://doi.org/10.1016/j.tjog.2024.04.005 (2024).

    Article  PubMed  Google Scholar 

  2. Stelzle D, Tanaka LF, Lee KK, Ibrahim Khalil A, Baussano I, Shah ASV, McAllister DA, Gottlieb SL, Klug SJ, Winkler AS, Bray F, Baggaley R, Clifford GM, Broutet N, Dalal S. Estimates of the global burden of cervical cancer associated with HIV. Lancet Glob Health. 9(2), e161-e169 (2021).

  3. Konwar, A. N. & Borse, V. Current status of point-of-care diagnostic devices in the Indian healthcare system with an update on COVID-19 pandemic. Sens. Int. 1, 100015. https://doi.org/10.1016/j.sintl.2020.100015 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  4. Chauhan, V. et al. Establishing national centre for point-of-care technology for “Swasth Bharat, Samridh Bharat”. J. Lab. Physicians 11(3), 177–179. https://doi.org/10.4103/JLP.JLP_139_19 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

  5. Muthuramalingam, M. R. & Muraleedharan, V. R. Patterns in the prevalence and wealth-based inequality of cervical cancer screening in India. BMC Womens Health 23(1), 337. https://doi.org/10.1186/s12905-023-02504-y (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Pimple, S. & Mishra, G. Cancer cervix: Epidemiology and disease burden. Cytojournal 19, 21. https://doi.org/10.25259/CMAS_03_02_2021 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

  7. Arrivillaga, M. et al. Women’s critical experiences with the pap smear for the development of cervical cancer screening devices. Heliyon 9(3), e14289. https://doi.org/10.1016/j.heliyon.2023.e14289 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  8. Perkins, R. B., Wentzensen, N., Guido, R. S. & Schiffman, M. Cervical cancer screening: A review. JAMA 330(6), 547–558. https://doi.org/10.1001/jama.2023.13174 (2023).

    Article  CAS  PubMed  Google Scholar 

  9. Strelow, B. et al. Saturday Pap Smear Clinic: Addressing barriers to cervical cancer screening. BMJ Open Qual. 12(3), e002252. https://doi.org/10.1136/bmjoq-2023-002252 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  10. Chu, M. M. et al. The value of four-quadrant cervical biopsy in women with different colposcopic impressions. Diagnostics (Basel) 13(14), 2384. https://doi.org/10.3390/diagnostics13142384 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  11. Ikanovic, M. et al. Fluorescence assay based on aptamer-quantum dot binding to Bacillus thuringiensis spores. J. Fluoresc. 17(2), 193–199. https://doi.org/10.1007/s10895-007-0158-4 (2007).

    Article  CAS  PubMed  Google Scholar 

  12. Deodhar, K. et al. Accuracy of concurrent visual and cytology screening in detecting cervical cancer precursors in rural India. Int. J. Cancer 131(6), E954–E962 (2012).

    Article  CAS  PubMed  Google Scholar 

  13. Hajian-Tilaki, K. Sample size estimation in diagnostic test studies of biomedical informatics. J. Biomed. Inform. 48, 193–204 (2014).

    Article  PubMed  Google Scholar 

  14. Killeen, J. L., Dye, T., Grace, C. & Hiraoka, M. Improved abnormal Pap smear triage using cervical cancer biomarkers. J. Low Genit. Tract Dis. 18, 1–7. https://doi.org/10.1097/LGT.0b013e31828aeb39 (2014).

    Article  PubMed  Google Scholar 

  15. Stoscheck, C. M. Quantitation of protein. Methods Enzymol. 82, 50–69 (1990).

    Article  Google Scholar 

  16. Tsikouras, P. et al. Cervical cancer: Screening, diagnosis and staging. J. BUON. 21, 320–325 (2016).

    PubMed  Google Scholar 

  17. Yim, E. K. & Park, J. S. Biomarkers in cervical cancer. Biomark. Insights 7(1), 215–225 (2007).

    Google Scholar 

  18. Lakowicz, J. R. Principles of Fluorescence Spectroscopy 3rd edn. (Springer, 2006). https://doi.org/10.1007/978-0-387-46312-4.

    Book  Google Scholar 

  19. Bhatla, N. et al. Cervical cancer prevention efforts in India. Indian J. Gynecol. Oncol. 19(3), 41. https://doi.org/10.1007/s40944-021-00526-8 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  20. Srivastava, A. N., Misra, J. S., Srivastava, S., Das, B. C. & Gupta, S. Cervical cancer screening in rural India: Status & current concepts. Indian J. Med. Res. 148(6), 687–696. https://doi.org/10.4103/ijmr.IJMR_5_17 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  21. Ferlay, J. et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int. J. Cancer 136(5), E359–E386. https://doi.org/10.1002/ijc.29210 (2015).

    Article  CAS  PubMed  Google Scholar 

  22. Man, I., Georges, D., Bonjour, M. & Baussano, I. Approximating missing epidemiological data for cervical cancer through footprinting: A case study in India. Elife 12, e81752. https://doi.org/10.7554/eLife.81752 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Kaur, S. et al. Challenges in cervical cancer prevention: Real-world scenario in India. South Asian J. Cancer 12(1), 9–16. https://doi.org/10.1055/s-0043-1764222 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  24. Singh, D. et al. Global estimates of incidence and mortality of cervical cancer in 2020: A baseline analysis of the WHO Global Cervical Cancer Elimination Initiative. Lancet Glob. Health 11(2), e197–e206. https://doi.org/10.1016/S2214-109X(22)00501-0 (2023).

    Article  CAS  PubMed  Google Scholar 

  25. Sung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 71, 209–249. https://doi.org/10.3322/caac.21660 (2020).

    Article  CAS  Google Scholar 

  26. World Health Organization. GLOBOCAN 2012: Cervical Cancer Estimated Incidence, Mortality and Prevalence Worldwide in 2012. GLOBOCON 2012

  27. Wu, J. et al. Global burden of cervical cancer: Current estimates, temporal trend and future projections based on the GLOBOCAN 2022. J. Natl. Cancer Center 5(3), 322–329. https://doi.org/10.1016/j.jncc.2024.11.006 (2025).

    Article  Google Scholar 

  28. Liu, Y. & Zheng, W. Cervical cancer development, screening, and prevention. In Gynecologic and Obstetric Pathology (eds Zheng, W. et al.) (Springer, 2025). https://doi.org/10.1007/978-981-97-4822-8_52.

    Chapter  Google Scholar 

  29. Locklar, L. R. B. & Do, D. P. Rural-urban differences in HPV testing for cervical cancer screening. J. Rural Health 38(2), 409–415. https://doi.org/10.1111/jrh.12615 (2022).

    Article  PubMed  Google Scholar 

  30. Srinivas, V. et al. Community-based mobile cervical cancer screening program in rural India: Successes and challenges for implementation. Asian Pac. J. Cancer Prev. 22(5), 1393–1400. https://doi.org/10.31557/APJCP.2021.22.5.1393 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  31. Sawaya, G. F. & Smith-McCune, K. Cervical cancer screening. Obstet. Gynecol. 127(3), 459–467. https://doi.org/10.1097/AOG.0000000000001136 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  32. Origoni, M. et al. HPV-DNA testing for cervical cancer precursors: from evidence to clinical practice. E Cancer Med. Sci. 6, 258. https://doi.org/10.3332/ecancer.2012.258 (2012).

    Article  CAS  Google Scholar 

  33. Salazar-Campos, J. E. et al. Cervicouterine cancer screening-TruScreen™ vs. conventional cytology: Pilot study. J. Cytol. 35(3), 143–148. https://doi.org/10.4103/JOC.JOC_111_17 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Tanaka, Y. et al. “Smartscopy” as an alternative device for cervical cancer screening: A pilot study. BMJ Innov. 3(2), 123–126. https://doi.org/10.1136/bmjinnov-2016-000174 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

  35. Jayabalan, S., Subbaiah, M. & Chaturvedula, L. Diagnostic accuracy of hand-held colposcope (Gynocular) in comparison with standard colposcope in patients with abnormal cervical cytology or visual inspection with acetic acid positivity: A cross over randomized controlled study. Obstet. Gynecol. Sci. 66(4), 300–306. https://doi.org/10.5468/ogs.23089 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  36. Peterson, C. W., Rose, D., Mink, J. & Levitz, D. Real-time monitoring and evaluation of a visual-based cervical cancer screening program using a decision support job aid. Diagnostics (Basel) 6(2), 20. https://doi.org/10.3390/diagnostics6020020 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Quinn, M. K. et al. High-resolution microendoscopy for the detection of cervical neoplasia in low-resource settings. PLoS ONE 7(9), e44924. https://doi.org/10.1371/journal.pone.0044924 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Coole, J. B. et al. Development of a multimodal mobile colposcope for real-time cervical cancer detection. Biomed. Opt. Express 13(10), 5116–5130. https://doi.org/10.1364/BOE.463253 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

  39. Sung, J., Wang, Y., Chandrasekaran, S., Witten, D. M. & Price, N. D. Molecular signatures from omics data: From chaos to consensus. Biotechnol. J. 7, 946–957. https://doi.org/10.1002/biot.201100305 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Heinzel, A. et al. From molecular signatures to predictive biomarkers: Modeling disease pathophysiology and drug mechanism of action. Front. Cell Dev. Biol. 2, 37. https://doi.org/10.3389/fcell.2014.00037 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  41. Bobdey, S., Sathwara, J., Jain, A. & Balasubramaniam, G. Burden of cervical cancer and role of screening in India. Indian J. Med. Paediatr. Oncol. 37, 278–285. https://doi.org/10.4103/0971-5851.195751 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  42. Srivastava, A. N., Misra, J. S., Srivastava, S., Das, B. C. & Gupta, S. Cervical cancer screening in rural India: Status & current concepts. Indian J. Med. Res. 148, 687–696. https://doi.org/10.4103/ijmr.IJMR_5_17 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  43. Mallath, M. K. et al. The growing burden of cancer in India: Epidemiology and social context. Lancet Oncol. 15, e205–e212. https://doi.org/10.1016/S1470-2045(14)70115-9 (2014).

    Article  PubMed  Google Scholar 

  44. Fujita, M., Yamada, C. & Goto, H. Cell cycle regulation of human Cdc 6 protein. J. Biol. Chem. 274, 25927–25932 (1999).

    Article  CAS  PubMed  Google Scholar 

  45. Williams, G. H. & Coleman, N. Minichromosome maintenance proteins as biological markers of dysplasia and malignancy. Clin. Cancer Res. 5, 2121–2132 (1999).

    PubMed  Google Scholar 

  46. Murphy, N. et al. p16INK4A as a marker for cervical dyskaryosis: CIN and cGIN in cervical biopsies and ThinPrep™ smears. J. Clin. Pathol. 56, 53–63 (2003).

    Article  Google Scholar 

  47. Klaes, R. et al. Overexpression of p16 as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int. J. Cancer 92, 276–284 (2001).

    Article  CAS  PubMed  Google Scholar 

  48. Philp, L. et al. Pap tests in the diagnosis of cervical cancer: Help or hinder?. Gynecol. Oncol. 150, 61–66. https://doi.org/10.1016/j.ygyno.2018.05.019 (2018).

    Article  PubMed  Google Scholar 

  49. Dijkstra, M. G. et al. Cervical cancer screening: On the way to a shift from cytology to full molecular screening. Ann. Oncol. 25, 927–935 (2014).

    Article  CAS  PubMed  Google Scholar 

  50. Origoni, M., Prendiville, W. & Paraskevaidis, E. Cervical cancer prevention: New frontiers of diagnostic strategies. Biomed. Res. Int. 8, 250917 (2015).

    Google Scholar 

  51. Petignat, P. & Roy, M. Diagnosis and management of cervical cancer. BMJ 335, 765–768. https://doi.org/10.1136/bmj.39337.615197.80 (2007).

    Article  PubMed  PubMed Central  Google Scholar 

  52. Vilos, G. A. The history of the Papanicolaou smear and the odyssey of George and Andromache Papanicolaou. Obstet. Gynecol. 91(3), 479–483. https://doi.org/10.1016/s0029-7844(97)00695-9 (1998).

    Article  CAS  PubMed  Google Scholar 

  53. Waldmann, A., Eisemann, N. & Katalinic, A. Epidemiology of malignant cervical, corpus uteri and ovarian tumours-current data and epidemiological trends. GeburtshilfeFrauenheilkd 73(2), 123–129. https://doi.org/10.1055/s-0032-1328266 (2013).

    Article  CAS  Google Scholar 

  54. Zhang, H. et al. CD146 is a potential marker for the diagnosis of malignancy in cervical and endometrial cancer. Oncol. Lett. 5, 1189–1194. https://doi.org/10.3892/ol.2013.1147 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Cheng, H. Inhibiting CD146 by its monoclonal antibody AA98 improves radio sensitivity of cervical cancer cells. Med. Sci. Monit. 20, 3328–3333. https://doi.org/10.12659/msm.896731 (2016).

    Article  CAS  Google Scholar 

  56. Güzel, C. et al. Molecular markers for cervical cancer screening. Expert Rev Proteomics 18(8), 675–691. https://doi.org/10.1080/14789450.2021.1980387 (2021).

    Article  CAS  PubMed  Google Scholar 

  57. Sarandakou, A. et al. Vaginal fluid and serum CEA, CA125 and SCC in normal conditions and in benign and malignant diseases of the genital tract. Acta Oncol. 36(7), 755–759. https://doi.org/10.3109/02841869709001350 (1997).

    Article  CAS  PubMed  Google Scholar 

  58. Honda, K. The biological role of actinin-4 (ACTN4) in malignant phenotypes of cancer. Cell Biosci. 5, 41. https://doi.org/10.1186/s13578-015-0031-0 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Ruiz Esparza Garrido, R., Gutiérrez, M. & Ángel Velázquez Flores, M. Circulating cervical cancer biomarkers potentially useful in medical attention (Review). Mol. Clin. Oncol. 18(2), 13. https://doi.org/10.3892/mco.2023.2609 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Shanta, V., Krishnamurthi, S., Gajalakshmi, C. K., Swaminathan, R. & Ravichandran, K. Epidemiology of cancer of the cervix: Global and national perspective. J. Indian Med. Assoc. 98, 49–52 (2000).

    CAS  PubMed  Google Scholar 

  61. van der Watt, P. J. et al. Nuclear transport proteins are secreted by cancer cells and identified as potential novel cancer biomarkers. Int. J. Cancer 150(2), 347–361. https://doi.org/10.1002/ijc.33832 (2022).

    Article  CAS  PubMed  Google Scholar 

  62. Murali, R. et al. Cytological features of transitional cell metaplasia of the lower female genital tract. Pathology 42(2), 113–118. https://doi.org/10.3109/00313020903494110 (2010).

    Article  PubMed  Google Scholar 

  63. Shin, E., Yu, J. & Hong, S. W. Trouble-makers in cytologic interpretation of the uterine cervix. J. Pathol. Transl. Med. 57(3), 139–146. https://doi.org/10.4132/jptm.2023.04.25 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

  64. Imai, Y., Fukagawa, Y., Sugihara, S. & Teranishi, T. Reclassification of atypical immature metaplasia of the uterine cervix by combination of nuclear features on hematoxylin and eosin–stained sections without auxiliary immunohistochemistry. Hum. Pathol. 129, 113–122. https://doi.org/10.1016/j.humpath.2022.09.002 (2022).

    Article  CAS  PubMed  Google Scholar 

  65. Bonds, L. et al. Immunohistochemical localization of CDC6 in squamous and glandular neoplasia of the uterine cervix. Arch. Pathol. Lab. Med. 126, 1162–1168 (2002).

    Article  Google Scholar 

  66. de Kruijff, I. E. et al. The prevalence of CD146 expression in breast cancer subtypes and its relation to outcome. Cancers 2018(10), 134. https://doi.org/10.3390/cancers10050134 (2018).

    Article  CAS  Google Scholar 

  67. Jiang, G. et al. CD146 promotes metastasis and predicts poor prognosis of hepatocellular carcinoma. J. Exp. Clin. Cancer Res. 29, 38. https://doi.org/10.1186/s13046-016-0313-3 (2016).

    Article  CAS  Google Scholar 

  68. Zheng, C. et al. Endothelial CD146 is required for in vitro tumor-induced angiogenesis: The role of a disulfide bond in signaling and dimerization. Int. J. Biochem. Cell Biol. 41, 2163–2172 (2009).

    Article  CAS  PubMed  Google Scholar 

  69. Liverani, C. A., Di Giuseppe, J., Giannella, L., Delli Carpini, G. & Ciavattini, A. Cervical cancer screening guidelines in the postvaccination era: Review of the literature. J. Oncol. https://doi.org/10.1155/2020/8887672 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  70. Mukherjee, A. G. et al. Exploring the molecular pathogenesis, pathogen association, and therapeutic strategies against HPV infection. Pathogens 12, 25 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

  71. Petignat, P. et al. Age-related performance of human papillomavirus testing used as an adjunct to cytology for cervical carcinoma screening in a population with a low incidence of cervical carcinoma. Cancer 105, 126–132 (2005).

    Article  PubMed  Google Scholar 

  72. Larkins, M. C. & Thombare, A. Point-of-Care Testing. [Updated 2023 May 29]. In Stat Pearls [Internet]. Treasure Island (FL): Stat Pearls Publishing; 2023 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK592387/

  73. Neogi, S. B. et al. Diagnostic accuracy of point-of-care devices for detection of anemia in community settings in India. BMC Health Serv. Res. 20(1), 468. https://doi.org/10.1186/s12913-020-05329-9 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

  74. Bujang, M. A. & Adnan, T. H. Requirements for minimum sample size for sensitivity and specificity analysis. J. Clin. Diagn. Res. 10(10), YE01–YE06 (2016). https://doi.org/10.7860/JCDR/2016/18129.8744.

  75. Bachmann, L. M., Puhan, M. A., Riet, G. & Bossuyt, P. M. Sample sizes of studies on diagnostic accuracy: Literature survey. BMJ 332, 1127. https://doi.org/10.1136/bmj.38793.637789.2F (2006).

    Article  PubMed  PubMed Central  Google Scholar 

  76. Indian Council of Medical Research. Consensus document for the management of cancer cervix [Internet]. New Delhi: ICMR; 2016 [cited 2026 Sep 22]. Available from: icmr.gov.in

  77. World Health Organization. Cervical cancer screening in developing countries: report of a WHO consultation [Internet]. Geneva: World Health Organization; 2002 [cited 2026 Sep 22]. Available from: https://who.int

  78. World Health Organization. WHO guidelines for screening and treatment of precancerous lesions for cervical cancer prevention [Internet]. Geneva: World Health Organization; 2013 [cited 2026 Sep 22]. Available from: https://who.int

  79. Mathew A, George PS. Trends in incidence and mortality rates of squamous cell carcinoma and adenocarcinoma of cervix–worldwide. Asian Pac. J. Cancer Prev. 10(4), 645-50 (2009).

  80. Zhao X, Kong W, Xie Y, Chotoo RA, Wang Y. Artificial Intelligence in Cervical Cancer Screening and Diagnosis. Front Oncol. 12, 851367 (2022).

  81. Principles of fluorescence spectroscopy by Lakowiez (1999). https://doi.org/10.1007/978-0-387-46312-4

  82. Cervical cancer overview, World Health Organization (WHO). https://www.who.int/health-topics/cervical-cancer#tab=tab_1

  83. Nanda, K. et al. Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: A systematic review. Ann. Intern. Med. 132(10), 810–819. https://doi.org/10.7326/0003-4819-132-10-200005160-00009 (2000).

    Article  CAS  PubMed  Google Scholar 

  84. Holtman, A. et al. Development of practical recommendations for diagnostic accuracy studies in low-prevalence situations. J. Clin. Epidemiol. 114, 38–48 (2019).

    Article  PubMed  Google Scholar 

  85. Yeşilfidan D, Özkan Öndeş M, Adana F. Comparison of cervical cancer-related knowledge, attitudes, and behaviors among women in urban and rural regions. BMC Public Health. 25(1), 2561 (2025).

  86. Khanna D, Bery A, Sharma A, Sharma S. Evaluating knowledge regarding cervical cancer and its screening among woman in rural India. J. Family Med. Primary Care. 10(2), 934-939 (2021).

  87. Godoy-Vitorino F, Romaguera J, Zhao C, Vargas-Robles D, Ortiz-Morales G, Vázquez-Sánchez F, Sanchez-Vázquez M, de la Garza-Casillas M, Martinez-Ferrer M, White JR, Bittinger K, Dominguez-Bello MG and Blaser MJ. Cervicovaginal Fungi and Bacteria Associated With Cervical Intraepithelial Neoplasia and High-Risk Human Papillomavirus Infections in a Hispanic Population. Front. Microbiol. 9, 2533 (2018).

  88. Di Pietro M, Filardo S, Porcella G, Avitabile E, Sessa R. HPV/Chlamydia trachomatis co-infection: metagenomic analysis of cervical microbiota in asymptomatic women. New Microbiol. 41(1), 34-41 (2018).

  89. Zhang X, Zeng Q, Cai W, Ruan W. Trends of cervical cancer at global, regional, and national level: data from the Global Burden of Disease study 2019. BMC Public Health [Internet]. 21(1), 894 (2021).

  90. Zhao, F. et al. The application of liquid-based cytological detection for P16, cytologic evaluation and high-risk human papillomavirus testing in cervical cancer screening: A clinical evaluation. Int. J. Womens Health 28(14), 965–973. https://doi.org/10.2147/IJWH.S365236 (2022).

    Article  Google Scholar 

Download references

来源:Nature 健康护理主题 · nature.com