Background and objectives. The Kingdom of Eswatini (Southern Africa) has the greatest cervical cancer (CC) burden worldwide. Socio-economic and cultural factors hinder access to CC screening and retention in care. Test invasiveness and sampling modalities are known to further reduce women's participation in screening programs. In line with the World Health Organisation's goal of eliminating cervical cancer as a global public health problem by the end of the century, the EpiSoMI and MACH Coordinated Research Centres of the University of Milan, in collaboration with Cabrini Ministries and Missionary Sisters of the Sacred Heart of Jesus and the support of Cepheid (California, US), introduced an innovative urine-based human papillomavirus (HPV)-DNA screening test in a small rural healthcare facility in Eswatini to increase access to, and acceptability of, CC prevention among adolescents and women. This thesis reports data regarding the feasibility, effectiveness, and cost of urine-based CC screening collected during the project’s pilot phase. Additionally, it aims to collect data on HPV infection prevalence and genotypes distribution among women and adolescents and to determine the potential impact of current vaccines on preventing these infections. Methods. We performed a 12-month-long cross-sectional pilot study (23 January 2023 – 29 February 2024) in which women aged 12-49 years self-presenting at St. Philip's Clinic were asked to provide 30 mL of urine. A 10 mL urine sample was centrifuged to concentrate viral particles and eliminate debris. The sample was tested locally with the Xpert® HPV test (Cepheid) to detect 14 high-risk (HR)-HPV genotypes. Women aged 21-49 were also offered cervical brush sampling for the Xpert® HPV test performed at St. Philip's and cytological analyses performed at Mbabane Central Pathology Laboratory (embedded study). A double-blinded reading of the digital Pap smear digital scan is currently being performed at the University of Milan. Qualitative analysis was used to define the key steps and issues of the proposed cervical cancer screening program. The performance of the HPV DNA test on urine samples was evaluated through the agreement against the HPV DNA test on cervical samples and the Pap test. Statistical analysis included descriptive statistics, chi-square tests, logistic regression, and Kappa statistics. HPV genotype distribution among women visiting St. Philip's Clinic was performed by characterising HR and low-risk (LR) HPV types via an Ampliquality HPV-type express kit (AB Analitica) and Sanger sequencing of the L1 region from Dried Urine Spot samples (DUSs) prepared at St. Philip’s and shipped to our laboratory at the University of Milan. Project set up and St. Philip's clinic staff training. The project was set up through two field visits and e-meetings, which were essential to organise the laboratory, train the nurse and laboratory staff involved in the screening project, and launch the project. Two additional field visits were undertaken to monitor the field laboratory activities during the project. Throughout the pilot phase, partners' and stakeholders' engagement has been critical to informing and updating Eswatini policymakers. Main results obtained within the urine-based HPV DNA screening for cervical cancer at St. Philip's clinic. During the pilot phase of the study, 512 women were enrolled (median age 29 years). Five hundred ten urine samples were collected, and 42% were HR-HPV positive. HIV-positive participants were nearly twice as likely to test positive for HR-HPV compared to HIV-negative. Cervical brush samples were collected for 220 women consecutively recruited at the start of the project. Two hundred nineteen were tested for HPV with Xpert®, with 36% of the samples showing positive results. Pap tests revealed abnormal cytology in 34% of the women enrolled in the embedded study. Notably, CC was identified in 5 women, 2 of which aged under 30 years old. Assessment of the screening programme feasibility. The implementation faced numerous challenges (e.g., only one nurse was available to handle all the tasks related to the screening and logistical issues such as mislabelled samples, errors in processing, and occasional equipment failures). Each issue was carefully investigated to find the best possible solution. Error and invalid results on the Xpert® HPV DNA on fresh urine samples determined an over-testing rate of 17.3%, with 88 out of 510 samples tested more than once. This number exceeded cervical brush samples (233 tests for 219 samples). The Xpert HPV DNA test on urine and cervix brush samples showed an agreement rate of 79% (Cohen's K 0.6). All samples collected from CC women tested positive for urine and cervix brush samples. The overall cost for the project's first phase was € 77920.7, including the cost of the laboratory equipment and consumables, the cost of the embedded study, and the cost of the staff directly involved in the screening. The price for screening for single women enrolled was € 8.6 for the Xpert HPV test on fresh urine, € 11 for the Xpert HPV test on the cervical brush, and € 36 for the Pap test. Molecular epidemiology study. Four hundred forty DUS were sent to the University of Milan laboratory, and 336 samples tested positive for HPV DNA (76%). Three hundred thirty-three samples were genotyped, with most testing positive for multiple genotypes (63%). HPV35 was the type detected more frequently (24%), followed by HPV16 (18%), HPV70 (15%), HPV52 (15%), and HPV31 HPV6 (13%). At least one HR-HPV type was detected in 84% of samples. The overall prevalence of HPV infection among women with cytological abnormalities was 79%, which was considerably higher than those classified as normal cytology (77%, p-value: 0.012). Infections sustained by HR-HPV genotypes included in the 9-valent anti-HPV vaccine (HPV16, 18, 31, 33, 45, 52, and 58) were detected in 22 women with high-grade cervical lesions (76%). The proportion was reduced to 41% (12/29), considering only HPV16 and 18 types (2-valent and 4-valent vaccine). Discussion. Data collected within the pilot phase of the project "Prevention of Cervical Cancer in a Rural Primary Care Centre in Eswatini" suggest that urine-based HPV screening can be effectively implemented in rural settings, providing an accessible and less invasive method for the early detection of cervical cancer. The pilot program enrolled 512 women in 1 year, revealing community interest. A high prevalence of HPV infection (42%) was observed in the project area, with abnormal cytology affecting also very young women, underlining the need for screening strategies involving the youngest. The pilot phase successfully highlighted the operational challenges, including infrastructural limitations, social barriers, and the need for continuous community engagement. However, practical solutions were identified for almost all such challenges. The fresh urine-based test showed promising results with moderate agreement with the cervix brush sample. Cost analysis suggests urine-based testing could be cheaper than traditional methods, but further validation is needed to ensure its long-term feasibility. To the best of our knowledge, this is the first study assessing the distribution of HPV genotypes in Eswatini. The high prevalence of HR-HPV genotypes, such as HPV35, raises concerns as current HPV vaccines do not target them, nor do most of the available rapid molecular diagnostics identify them individually. Evaluation of their impact on the development of cervical high-grade lesions and cancer awaits further studies. Our preliminary findings can inform national and international policymakers and implementers in developing and adopting cost-effective prevention strategies for cervical cancer prevention. Overall, the project marks an important first step forward in cervical cancer prevention in Eswatini, underlining the importance of tailored approaches to healthcare challenges in low—and middle-income countries.
Contesto e Obiettivi. Il Regno di Eswatini (Africa meridionale) presenta la più alta incidenza e mortalità per cancro cervicale (CC) a livello globale. Fattori socio-economici e culturali ostacolano l'accesso ai programmi di screening per il CC e la permanenza nel percorso di cura. L'invasività del test e le modalità di campionamento sono ulteriori barriere che riducono la partecipazione delle donne ai programmi di screening. In linea con l'obiettivo dell'Organizzazione Mondiale della Sanità di eliminare il cancro cervicale come problema di salute pubblica globale entro la fine del secolo, i centri di ricerca coordinata EpiSoMI e MACH dell'Università degli Studi di Milano, in collaborazione con Cabrini Ministries e le Suore Missionarie del Sacro Cuore di Gesù e con il contributo di Cepheid (California, US), hanno introdotto un test di screening basato su un human papillomavirus (HPV) DNA test in campioni di urina in una piccola clinica rurale di Eswatini, al fine di aumentare l'accesso e l'accettabilità della prevenzione del CC tra le adolescenti e le donne. Questa tesi riporta i risultati della fase pilota del progetto, con l'obiettivo di stabilire la fattibilità, l'efficacia e i costi dell'utilizzo di campioni di urina per lo screening del CC. Inoltre, mira a raccogliere dati sulla prevalenza delle infezioni da HPV e sulla distribuzione dei genotipi tra donne e adolescenti, e a valutare l'impatto potenziale dei vaccini attuali nella prevenzione di tali infezioni. Metodi. Dal 23 gennaio 2023 al 29 febbraio 2024 è stato condotto uno studio pilota trasversale durante il quale le donne di età compresa tra 12 e 49 anni afferenti alla clinica di St. Philip’s sono state invitate a fornire 30 mL di urina. Un campione di 10 mL di urina è stato centrifugato per concentrare le particelle virali ed eliminare i detriti, e il campione è stato testato localmente con il test Xpert® HPV (Cepheid) per rilevare 14 genotipi di HPV ad alto rischio oncogeno. Alle donne di età compresa tra 21 e 49 anni è stato inoltre proposto un secondo campionamento tramite brush cervicale per il test Xpert® HPV, eseguito presso il laboratorio della clinica di St. Philip’s, e per le analisi citologiche eseguite presso il laboratorio Centrale di Patologia di Mbabane (embedded study). Una doppia lettura in cieco di scansioni digitali degli strisci citologici effettuati al laboratorio di Patologia di Mbabane è al momento in corso presso l’Università degli Studi di Milano. Tramite un’analisi quantitativa sono stati definiti i passaggi chiave e le problematiche del programma di screening proposto. La performance dell’HPV-DNA test sui campioni di urina è stata valutata confrontandola con l’HPV-DNA test effettuato su brush cervicali e con il Pap test. L'analisi statistica ha incluso statistiche descrittive, test del chi-quadrato, regressione logistica e statistiche Kappa. Tramite la caratterizzazione dei genotipi di HPV ad alto e basso rischio oncogeno con tecniche di biologia molecolare quali Ampliquality HPV-type express (AB Analitica) e il sequenziamento Sanger della regione L1 è stata descritta la distribuzione dei genotipi di HPV diffusi tra le donne afferenti alla clinica di St. Philip’s. I test sono stati condotti su campioni di urine essiccate (DUS) preparati a St. Philip’s e inviati al nostro laboratorio dell'Università di Milano. Avvio del progetto e formazione del personale della clinica di St. Philip’s. L’organizzazione del progetto ha richiesto diversi incontri online e due visite sul campo, finalizzate ad organizzare il laboratorio, formare gli infermieri e i tecnici di laboratorio coinvolti nel progetto di screening e dare il via al progetto. Durante l’anno si sono verificate altre due visite in loco per monitorare le attività del laboratorio sul campo. Durante la fase pilota, il coinvolgimento dei partner e degli stakeholder è stato fondamentale per informare e aggiornare i policymakers di Eswatini. Risultati principali ottenuti nell'ambito dello screening HPV-DNA urinario per il cancro cervicale presso la clinica di St. Philip’s. Durante la fase pilota dello studio, sono state arruolate 512 donne (età mediana 29 anni). Di 510 campioni di urina raccolti, il 42% è risultato HR-HPV positivo. Le partecipanti HIV-positive hanno mostrato un rischio di risultare HPV DNA positive quasi doppio rispetto a quelle HIV-negative. Duecentoventi donne reclutate consecutivamente dall’inizio del progetto hanno partecipato allo studio fornendo anche il brush cervicale e 219 campioni sono stati testati per tramite Xpert® HPV test, con il 36% dei campioni risultati positivi. I Pap test hanno rivelato citologia anormale nel 34% delle donne arruolate nell’embedded study. Degno di nota, il Pap test ha rilevato la presenza di CC in 5 donne, di cui 2 di età inferiore a 30 anni. Valutazione della fattibilità del programma di screening. L'implementazione del programma di screening ha incontrato numerose sfide (ad esempio, solo un'infermiera era disponibile per gestire tutti i compiti relativi allo screening, ci sono stati problemi logistici di etichettature dei campioni, errori di elaborazione dei risultati e occasionali guasti delle apparecchiature). Tuttavia, ogni problema è stato approfondito e indagato per trovare la soluzione migliore possibile. Errori e risultati non validi nel Xpert® HPV test sui campioni di urina hanno determinato un tasso di test ripetuti del 17%, con 88 dei 510 campioni testati più di una volta. Questo numero ha superato quello dei brush cervicali (233 test per 219 campioni). L’Xpert® HPV test sulle due tipologie di campioni ha mostrato un tasso di concordanza del 79% (Cohen's K 0.6) e il campione di urina è riuscito ad identificare l’infezione da HPV in tutte le donne con CC. Il costo complessivo per la prima fase del progetto è stato di € 77.920,7, comprendendo i costi delle attrezzature di laboratorio e dei materiali di consumo, dell’embedded study e del personale direttamente coinvolto nello screening. Il costo dell’HPV DNA test è stato di € 8,6 per donna testando i campioni di urina e € 11 testando il brush cervicale. Il Pap test invece è costato € 36. Studio di epidemiologia molecolare. Quattrocentoquaranta DUS sono state inviate al laboratorio dell'Università di Milano e 336 sono risultate HPV-DNA positive (76%). Trecentotrentatre campioni sono stati genotipizzati e nel 63% è stata rilevata la presenza di infezioni multiple. HPV35 è stato il tipo rilevato più frequentemente (24%), seguito da HPV16 (18%), HPV70 (15%), HPV52 (15%) e HPV31 HPV6 (13%). Almeno un HR-HPV è stato rilevato nell'84% dei campioni. La prevalenza complessiva delle infezioni da HPV tra le donne con anomalie citologiche è stata del 79%, significativamente superiore rispetto a quelle classificate come citologia normale (77%, p-value: 0,012). Le infezioni sostenute da HR-HPV inclusi nel vaccino anti-HPV 9-valente (HPV16, 18, 31, 33, 45, 52 e 58) sono state rilevate in 22 donne con lesioni cervicali di alto grado (76%). La proporzione si riduce al 41% (12/29) considerando solo i tipi HPV16 e 18 (vaccino bi- e quadrivalente). Discussione. I dati raccolti suggeriscono che il test di screening basato su HPV DNA test in campioni di urina può essere implementato efficacemente in contesti rurali, fornendo un metodo accessibile e meno invasivo per la diagnosi precoce del cancro cervicale. Il programma pilota ha arruolato 512 donne in 1 anno, rivelando un forte interesse da parte della comunità. È stata osservata un'alta prevalenza di infezioni da HPV (42%) nell’area di interesse del progetto, con citologia anormale anche in donne molto giovani, sottolineando la necessità di strategie di screening che coinvolgano le adolescenti. La fase pilota si è rilevata essenziale per mettere in luce le sfide operative di un programma di screening basato su HPV-DNA test in una clinica rurale dell’ultimo miglio. Soluzioni pratiche sono state individuate per tutti i problemi riscontrati, dimostrando la fattibilità operativa del programma. I test sulle urine hanno mostrato risultati promettenti, con una concordanza moderata rispetto al brush cervicale. L'analisi dei costi suggerisce che lo screening su campioni di urina potrebbe essere più economico rispetto ai metodi tradizionali, ma è necessaria un'ulteriore validazione per garantirne la fattibilità a lungo termine. Questo è il primo studio che valuta la distribuzione dei genotipi di HPV in Eswatini. L'alta prevalenza di genotipi HR-HPV non inclusi nelle formulazioni vaccinali ad oggi disponibili, come HPV35, è in linea con quanto riportato per altri LMICs e suscita preoccupazione e mette in luce la necessità di studi approfonditi per valutare il peso di questi genotipi nello sviluppo di lesioni di alto grado e cancro a livello locale. I nostri risultati preliminari possono fornire indicazioni ai decisori politici nazionali ed internazionali e agli operatori nell'elaborare e adottare strategie di prevenzione del cancro cervicale economicamente sostenibili e adatte alla realtà locale. Nel complesso, il progetto rappresenta un importante passo avanti nella prevenzione del cancro cervicale in Eswatini, sottolineando l'importanza di approcci personalizzati per affrontare le sfide sanitarie nei paesi a basso e medio reddito.
PREVENTION OF CERVICAL CANCER IN A RURAL PRIMARY CARE CENTER IN ESWATINI / C. Fappani ; tutor: E. Tanzi ; coordinatore: C. La Vecchia. Dipartimento di Scienze della Salute, 2025 Jan 28. 37. ciclo, Anno Accademico 2023/2024.
PREVENTION OF CERVICAL CANCER IN A RURAL PRIMARY CARE CENTER IN ESWATINI
C. Fappani
2025
Abstract
Background and objectives. The Kingdom of Eswatini (Southern Africa) has the greatest cervical cancer (CC) burden worldwide. Socio-economic and cultural factors hinder access to CC screening and retention in care. Test invasiveness and sampling modalities are known to further reduce women's participation in screening programs. In line with the World Health Organisation's goal of eliminating cervical cancer as a global public health problem by the end of the century, the EpiSoMI and MACH Coordinated Research Centres of the University of Milan, in collaboration with Cabrini Ministries and Missionary Sisters of the Sacred Heart of Jesus and the support of Cepheid (California, US), introduced an innovative urine-based human papillomavirus (HPV)-DNA screening test in a small rural healthcare facility in Eswatini to increase access to, and acceptability of, CC prevention among adolescents and women. This thesis reports data regarding the feasibility, effectiveness, and cost of urine-based CC screening collected during the project’s pilot phase. Additionally, it aims to collect data on HPV infection prevalence and genotypes distribution among women and adolescents and to determine the potential impact of current vaccines on preventing these infections. Methods. We performed a 12-month-long cross-sectional pilot study (23 January 2023 – 29 February 2024) in which women aged 12-49 years self-presenting at St. Philip's Clinic were asked to provide 30 mL of urine. A 10 mL urine sample was centrifuged to concentrate viral particles and eliminate debris. The sample was tested locally with the Xpert® HPV test (Cepheid) to detect 14 high-risk (HR)-HPV genotypes. Women aged 21-49 were also offered cervical brush sampling for the Xpert® HPV test performed at St. Philip's and cytological analyses performed at Mbabane Central Pathology Laboratory (embedded study). A double-blinded reading of the digital Pap smear digital scan is currently being performed at the University of Milan. Qualitative analysis was used to define the key steps and issues of the proposed cervical cancer screening program. The performance of the HPV DNA test on urine samples was evaluated through the agreement against the HPV DNA test on cervical samples and the Pap test. Statistical analysis included descriptive statistics, chi-square tests, logistic regression, and Kappa statistics. HPV genotype distribution among women visiting St. Philip's Clinic was performed by characterising HR and low-risk (LR) HPV types via an Ampliquality HPV-type express kit (AB Analitica) and Sanger sequencing of the L1 region from Dried Urine Spot samples (DUSs) prepared at St. Philip’s and shipped to our laboratory at the University of Milan. Project set up and St. Philip's clinic staff training. The project was set up through two field visits and e-meetings, which were essential to organise the laboratory, train the nurse and laboratory staff involved in the screening project, and launch the project. Two additional field visits were undertaken to monitor the field laboratory activities during the project. Throughout the pilot phase, partners' and stakeholders' engagement has been critical to informing and updating Eswatini policymakers. Main results obtained within the urine-based HPV DNA screening for cervical cancer at St. Philip's clinic. During the pilot phase of the study, 512 women were enrolled (median age 29 years). Five hundred ten urine samples were collected, and 42% were HR-HPV positive. HIV-positive participants were nearly twice as likely to test positive for HR-HPV compared to HIV-negative. Cervical brush samples were collected for 220 women consecutively recruited at the start of the project. Two hundred nineteen were tested for HPV with Xpert®, with 36% of the samples showing positive results. Pap tests revealed abnormal cytology in 34% of the women enrolled in the embedded study. Notably, CC was identified in 5 women, 2 of which aged under 30 years old. Assessment of the screening programme feasibility. The implementation faced numerous challenges (e.g., only one nurse was available to handle all the tasks related to the screening and logistical issues such as mislabelled samples, errors in processing, and occasional equipment failures). Each issue was carefully investigated to find the best possible solution. Error and invalid results on the Xpert® HPV DNA on fresh urine samples determined an over-testing rate of 17.3%, with 88 out of 510 samples tested more than once. This number exceeded cervical brush samples (233 tests for 219 samples). The Xpert HPV DNA test on urine and cervix brush samples showed an agreement rate of 79% (Cohen's K 0.6). All samples collected from CC women tested positive for urine and cervix brush samples. The overall cost for the project's first phase was € 77920.7, including the cost of the laboratory equipment and consumables, the cost of the embedded study, and the cost of the staff directly involved in the screening. The price for screening for single women enrolled was € 8.6 for the Xpert HPV test on fresh urine, € 11 for the Xpert HPV test on the cervical brush, and € 36 for the Pap test. Molecular epidemiology study. Four hundred forty DUS were sent to the University of Milan laboratory, and 336 samples tested positive for HPV DNA (76%). Three hundred thirty-three samples were genotyped, with most testing positive for multiple genotypes (63%). HPV35 was the type detected more frequently (24%), followed by HPV16 (18%), HPV70 (15%), HPV52 (15%), and HPV31 HPV6 (13%). At least one HR-HPV type was detected in 84% of samples. The overall prevalence of HPV infection among women with cytological abnormalities was 79%, which was considerably higher than those classified as normal cytology (77%, p-value: 0.012). Infections sustained by HR-HPV genotypes included in the 9-valent anti-HPV vaccine (HPV16, 18, 31, 33, 45, 52, and 58) were detected in 22 women with high-grade cervical lesions (76%). The proportion was reduced to 41% (12/29), considering only HPV16 and 18 types (2-valent and 4-valent vaccine). Discussion. Data collected within the pilot phase of the project "Prevention of Cervical Cancer in a Rural Primary Care Centre in Eswatini" suggest that urine-based HPV screening can be effectively implemented in rural settings, providing an accessible and less invasive method for the early detection of cervical cancer. The pilot program enrolled 512 women in 1 year, revealing community interest. A high prevalence of HPV infection (42%) was observed in the project area, with abnormal cytology affecting also very young women, underlining the need for screening strategies involving the youngest. The pilot phase successfully highlighted the operational challenges, including infrastructural limitations, social barriers, and the need for continuous community engagement. However, practical solutions were identified for almost all such challenges. The fresh urine-based test showed promising results with moderate agreement with the cervix brush sample. Cost analysis suggests urine-based testing could be cheaper than traditional methods, but further validation is needed to ensure its long-term feasibility. To the best of our knowledge, this is the first study assessing the distribution of HPV genotypes in Eswatini. The high prevalence of HR-HPV genotypes, such as HPV35, raises concerns as current HPV vaccines do not target them, nor do most of the available rapid molecular diagnostics identify them individually. Evaluation of their impact on the development of cervical high-grade lesions and cancer awaits further studies. Our preliminary findings can inform national and international policymakers and implementers in developing and adopting cost-effective prevention strategies for cervical cancer prevention. Overall, the project marks an important first step forward in cervical cancer prevention in Eswatini, underlining the importance of tailored approaches to healthcare challenges in low—and middle-income countries.File | Dimensione | Formato | |
---|---|---|---|
phd_unimi_R13316.pdf
embargo fino al 17/07/2025
Descrizione: PhD thesis
Tipologia:
Altro
Dimensione
9.46 MB
Formato
Adobe PDF
|
9.46 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.