Urine-based microsatellite loss of heterozygosity as a pre-cytological biomarker for cervical cancer: two sequential pilot studies from Sudan

Abstract

Cervical cancer is the second most common malignancy in Sudanese women, with most cases presenting at an advanced stage because functional screening infrastructure is absent across the majority of the country. The Papanicolaou smear faces substantial cultural and logistical barriers; a large proportion of women decline speculum-based examination, particularly those who have undergone female genital cutting. Self-collected urine offers a biologically and culturally acceptable alternative matrix. Loss of heterozygosity (LOH) at chromosomal loci encoding tumor suppressor proteins is among the earliest molecular events in HPV-driven cervical carcinogenesis. We report two sequential proof-of-concept pilot studies, conducted in 2010 and 2016, investigating whether urinary DNA from Sudanese women carries detectable microsatellite LOH signals. Study 1 (2010) enrolled 20 women attending a routine gynecology clinic at Khartoum Teaching Hospital, Khartoum, Sudan. All had concurrent Pap smears assessed by an independent cytologist. Study 2 (2016) enrolled 23 women with histologically confirmed cervical carcinoma at the Radiation and Isotopes Centre, Khartoum (RICK), Sudan, with paired blood DNA as a germline reference. Both studies extracted DNA from urinary cell pellets by the Guanidine Chloride method and assessed LOH at four microsatellite markers —D3S1300 (3p14.2/ FHIT ), D3S1260 (3p22.2), D11S528 (11q23.3/ CADM1 ), and D11S35 (11q22.1/ ATM ) — by (Polymerase Chain Reaction) PCR, with products verified on 1.5% agarose gel and resolved by 8% polyacrylamide gel electrophoresis (PAGE). In Study 1, 16 of 20 women yielded no allele loss at any marker, consistent with their normal cytology. Critically, four women showed urinary allele-loss patterns consistent with LOH at D3S1300 (1/4) or D11S35 (3/4), yet all four had entirely normal or non-atypical concurrent Pap smears, indicating sub-cytological molecular detection. In Study 2, LOH was confirmed somatically, absent from all paired blood samples in 18 of 23 urine specimens (78.3%; 95% CI 56.3–92.5%). Individual marker frequencies were: D3S1300 52.2%, D11S528 39.1%, D3S1260 34.8%, D11S35 30.4%. Multi-locus LOH at two or more markers was present in 34.8% of cases. These proof-of-concept pilot studies, provide the first Sudanese evidence that urinary microsatellite LOH analysis is technically feasible and yields somatic-specific signals from cervical neoplastic epithelium. The four informative Study 1 specimens and the 23 Study 2 specimens are insufficient for definitive sensitivity or specificity estimation (95% CI for Study 2 detection rate: 56.3–92.5%); all findings are hypothesis-generating only. The sub-cytological allele-loss finding in Study 1 requires prospective validation with blood controls and longitudinal follow-up before any clinical inference is warranted. Failure to detect LOH in 21.7% of Study 2 cancers most likely reflects non-LOH mechanisms of tumor suppressor gene (TSG) silencing, principally promoter methylation, addressable by a combined LOH-methylation panel in future work.