Abstract Introduction Knowledge about the oncologic status of pelvic lymph nodes forms an essential and integral part in the management of women with uterine cancer. Lymph node status is part of endometrial cancer staging and plays an important role in primary treatment and adjuvant treatment planning and prognosis in women with cervical cancer. Current practice in the management of uterine cancers involves systematic full pelvic lymphadenectomy, mainly to determine the oncological status of the nodes, as there is no high-quality evidence suggesting a therapeutic effect attributable to lymphadenectomy. Imaging in the form of computed tomography (CT) scans and magnetic resonance (MRI) scan is not accurate to determine pelvic lymph node status in women with uterine cancer. Functional scans such as 18Fluoro-deoxy-glucose positron emission/computed tomography (FDG-PET/CT) scan might provide better access in this setting. Sentinel lymph node biopsy (SLNB) procedures, specifically the SLNB algorithm, have been proposed as a safe and accurate alternative procedure to full systematic lymphadenectomy in women with uterine cancers. It has also been proposed as a better alternative than complete omission of lymphadenectomy in women with presumed low risk early stage endometrial cancer. SLNB procedures might also be able to detect higher rates of lymph node metastases with the detection of micro metastases following pathological ultrastaging The presence or absence of high risk human papilloma virus (hrHPV) DNA in sentinel lymph nodes of women with cervical cancer has also been suggested to be a useful adjunct to frozen section examination (FSE) in assisting with determination of the status of the non-sentinel nodes. Some data suggest the combination of negative FSE and absence of hrHPV accurately predict the absence of metastases. South African women have high prevalence of human immunodeficiency virus infection, tuberculosis (TB) and pelvic inflammatory disease (PID). All these infections involve the lymphatic system. Data on SLNB procedures are form well-developed countries with different disease burdens and socioeconomic profiles, and there is no data from women living in low-resource settings. Aims This study aimed to determine the efficacy of and performance of FDG-PET/CT scan and SLNB and SLNB algorithm in accurately predicting the regional lymph node status of the pelvis in women with early stage cervical cancer and presumed early stage endometrial cancer. It also aimed to investigate the usefulness of HPV DNA testing of sentinel nodes in women with cervical cancer. Population and setting This was a prospective observational study performed in the Gynaecologic Oncology Unit at the Kalafong Provincial Tertiary Hospital and Steve Biko Academic Hospital. Patients aged 18 years and older, with operable stages cervical cancer and presumed early stage endometrial cancer willing and able to provide informed consent were eligible for inclusion. Materials and methods Sentinel node mapping was done using methylene blue (MB) and indocyanine green (ICG) injected into the cervix after induction of anaesthesia at the time of primary surgery. 99Technetium nanocolloid (99Tc) was administered one day pre-operatively followed by lymphoscintigram. FDG-PET/CT scans were performed prior to surgery. Following mapping and removal, FSE, HPV DNA typing, haematoxylin and eosin (H&E) examination with ultrastaging on H&E negative specimens were performed on the SLNs. All patients underwent systematic full pelvic lymphadenectomy and appropriate cancer surgery. Results One hundred patients were prospectively recruited to the study and results of 94 patients were available for analysis. SNL detection rate of the whole group was 60.6% with bilateral detection 29.2%. Twenty-four patients (25.5%) had pelvic metastases. Sixty-five percent of women with cervical cancer in this study were HIV positive, and the SLN detection rate in this group was 65% with bilateral detection rate of 30%. The detection rate was significantly higher in women without nodal metastases, those with stage IA2 – IB2 disease, with tumour less than 2 cm and women with BMI less than 25 kg/m2. HIV status, history of TB, PID and the presence of adhesions did not influence the SLN detection rate. The sentinel lymph node biopsy algorithm has a sensitivity of 100%, NPV of 100% and a false negative rate of 0% in this study. The SLNB procedure identified two women with only micro metastases (15.4%). These women would not have been identified with systematic lymphadenectomy and H&E examination. Indocyanine green and the combination of methylene blue and 99Technetium nanocolloid had significantly better sentinel node detection rates compared to methylene blue alone FDG-PET/CT scan was performed in 28 women. The sensitivity, specificity, positive and negative predictive values of FDG-PET/CT scans to accurately predict nodal status, were 66.67%, 82%, 30.77% and 95.38% respectively. The false negative rate of FDG-PET/CT scans was 33.3%. The sensitivity, specificity, PPV and NPV for FSE in this cohort was 66.67%, 100%, 100% and 96.05% respectively. The FNR for FSE was 23.1%. Thirty-two patients with cervical cancer had tumour and SLN hrHPV DNA data. The sensitivity, specificity, PPV and NPV of sentinel lymph node HPV DNA to predict metastases was 50%, 69.6%, 30 and 84.2% respectively with a false negative rate of 42.8%. Conclusions Although the SLN detection rate was lower compared to the published literature, the SLNB algorithm performed excellently in this group of patients of which the majority were HIV-infected. The SLNB procedure can be considered as a treatment option in selected cases in the management of women with early stage endometrial and cervical cancer. PET/CT should not be used as part of the primary diagnosis and staging investigations in women with uterine cancer, and is recommended only in selected cases for initial staging of locally advanced cervical cancer being considered for radical chemoradiation therapy. In this study, testing for the presence of hrHPV DNA in the sentinel lymph nodes was not useful as a predictor of pelvic lymph node status. The combination of negative FSE and negative hrHPV in the SLNs did not have a reliable negative predictive value for the absence of pelvic nodal metastases. / Thesis (PhD)--University of Pretoria, 2017. / Obstetrics and Gynaecology / PhD / Unrestricted