Timelines to Cervical Cancer Diagnosis and Treatment at a Tertiary Hospital in Botswana

Comparative study of machine learning and statistical survival models for enhancing cervical cancer prognosis and risk factor assessment using SEER data

Cervical cancer is a common malignant tumor of the female reproductive system and the leading cause of death among women worldwide. The survival prediction method can be used to effectively analyze the time to event, which is essential in any clinical study. This study aims to bridge the gap between traditional statistical methods and machine learning in survival analysis by revealing which techniques are most effective in predicting survival, with a particular emphasis on improving prediction accuracy and identifying key risk factors for cervical cancer. Women with cervical cancer diagnosed between 2013 and 2015 were included in our study using data from the Surveillance, Epidemiology, and End Results (SEER) database. Using this dataset, the study assesses the performance of Weibull, Cox proportional hazards models, and Random Survival Forests in terms of predictive accuracy and risk factor identification. The findings reveal that machine learning models, particularly Random Survival Forests (RSF), outperform traditional statistical methods in both predictive accuracy and the discernment of crucial prognostic factors, underscoring the advantages of machine learning in handling complex survival data. However, for a survival dataset with a small number of predictors, statistical models should be used first. The study finds that RSF models enhance survival analysis with more accurate predictions and insights into survival risk factors but highlights the need for larger datasets and further research on model interpretability and clinical applicability.

Diagnostic and pre-treatment intervals among patients with cervical cancer attending care at the Uganda Cancer Institute: a cross-sectional study

BACKGROUND

Majority of patients with cervical cancer in the low- and middle-income countries experience long diagnostic and pre-treatment intervals. This study sought to determine the factors associated with the diagnostic and pre-treatment intervals among patients with cervical cancer.

METHODS

This was a cross-sectional study conducted at the Uganda Cancer Institute (UCI) during October 2019 to January 2020. Patients aged ≥ 18 years with histological diagnosis of cervical cancer were consecutively sampled. Data were collected using a pre-tested semi-structured questionnaire and a data abstraction form. Diagnostic intervals, defined as the time between first visit of a patient to a primary healthcare provider to time of getting confirmed diagnosis, of ≤ 3 months was defined as early & >3 months as late. Pre-treatment intervals, which is the time from histological diagnosis to starting cancer chemo-radiotherapy of ≤ 1 month was defined as early and > 1 month as late. Data were analysed using STATA version 14.0. We used modified Poisson regression models with robust variance to determine socio-demographic and clinical factors associated with the intervals.

RESULTS

The mean age of the participants was 50.0 ± 11.7 years. The median diagnostic and pre-treatment intervals were 3.1 (IQR: 1.4-8.2) months and 2.4 (IQR: 1.2-4.1) months respectively. Half of the participants, 49.6% (200/403) were diagnosed early; one in 5 patients, 20.1% (81/403) promptly (within one month) initiated cancer chemo-radiotherapy. Participants more likely to be diagnosed early included those referred from district hospitals (level 5) (aPR = 2.29; 95%CI: 1.60-3.26) and with squamous cell carcinomas (aPR = 1.55; 95%CI: 1.07-2.23). Participants more likely to be diagnosed late included those who first discussed their symptoms with relatives, (aPR = 0.77; 95%CI: (0.60-0.98), had > 2 pre-referral visits (aPR = 0.75; 95%CI (0.61-0.92), and had advanced stage (stages 3 or 4) (aPR = 0.68; 95%CI: 0.55-0.85). Participants more likely to initiate cancer chemo-radiotherapy early included older patients (≥ 60 years) (aPR = 2.44; 95%CI: 1.18-5.03). Patients likely to start treatment late were those who had ≥2 pre-referral visits  (aPR = 0.63; 95%CI: 0.41-0.98) and those that took 3 - 6 months with symptoms before seeking healthcare (aPR = 0.52;95%CI: 0.29 - 0.95).

CONCLUSION

Interventions to promote prompt health-seeking and early diagnosis of cervical cancer need to target primary healthcare facilities and aim to enhance capacity of primary healthcare professionals to promptly initiate diagnostic investigations. Patients aged < 60 years require targeted interventions to promote prompt initiation of chemo-radiation therapy.

Access to Radiation Therapy and Related Clinical Outcomes in Patients With Cervical and Breast Cancer Across Sub-Saharan Africa: A Systematic Review

PURPOSE

To better understand the barriers to accessing standard-of-care radiation therapy (RT) for breast and cervical cancer in sub-Saharan Africa and their impact on outcomes.

METHODS

A comprehensive literature search was completed with a medical librarian. Articles were screened by title, abstract, and full text. Included publications were analyzed for data describing barriers to RT access, available technology, and disease-related outcomes, and further grouped into subcategories and graded according to predefined criteria.

RESULTS

A total of 96 articles were included: 37 discussed breast cancer, 51 discussed cervical cancer, and eight discussed both. Financial access was affected by health care system payment models and combined burdens of treatment-related costs and lost wages. Staffing and technology shortages limit the ability to expand service locations and/or increase capacity within existing centers. Patient factors including use of traditional healers, fear of stigma, and low health literacy decrease the likelihood of early presentation and completion of therapies. Survival outcomes are worse than most high- and middle-income countries and are affected by many factors. Side effects are similar to other regions, but these findings are limited by poor documentation capabilities. Access to palliative RT is more expeditious than definitive management. RT was noted to lead to feelings of burden, lower self-esteem, and worsened quality of life.

CONCLUSION

Sub-Saharan Africa represents a diverse region with barriers to RT that differ on the basis of funding, available technology and staff, and community populations. Although long-term solutions must focus on building capacity by increasing the number of treatment machines and providers, short-term improvements should be implemented, such as interim housing for traveling patients, increased community education to reduce late-stage diagnoses, and use of virtual visits to avoid travel.

Is postoperative radiotherapy (PORT) a viable option in high-risk early-stage cervical cancer after upfront or downstaged radical surgery? A comparative study

BACKGROUND

Radical surgery for cervical cancer has inherent benefits, and as upfront or post neoadjuvant chemotherapy (NACT), is extendable to locally advanced cancer cervix (LACC), with postoperative radiotherapy (PORT) for high-risk factors. Objective of the study was to compare the effectiveness and survival between non-PORT and PORT in high-risk early stages.

MATERIALS AND METHODS

Radical hysterectomies conducted between January 2014 and December 2017 were evaluated and followed till December 2019. Clinical, surgical-pathologic characteristics, and oncological outcomes were compared between non-PORT and PORT groups. A similar comparison was made between alive and dead patients within each group. The impact of PORT was assessed.

RESULTS

Of 178 radical surgeries, early-LACC constituted 70%. Most (37%) of the patients belonged to stage 1b2, while stage 2b formed 5%. Mean age of patients was 46.5 years; 69% were below 50 years of age. Abnormal bleeding (41%) was the predominant symptom, followed by postcoital (20%) and postmenopausal bleeding (12%). Upfront surgeries formed 70.2%, and the average waiting period was 1.93 months (range: 1-10 months). PORT patients were 97 (54.5%) in number and the remaining formed the non-PORT group. Mean follow-up was 34 months, with 118 (66%) alive patients. Significant adverse prognostic factors were tumors >4 cm (44.4% patients), positive margins (10%), lymphatic vascular space invasion (LVSI; 42%), malignant nodes (33%), multiple metastatic nodes averaging seven (range: 3-11), and delayed (>6 months) presentation, but not deep stromal invasion (77% patients) and positive parametrium (8.4% patients). PORT overcame the adverse effects of tumors >4 cm, multiple metastatic nodes, positive margins, and LVSI. Total recurrences (25%) were balanced for both groups, but recurrences within 2 years were significantly more for PORT. Two-year overall survival (78%) and recurrence-free survival (72%), median overall survival (21 months), and median recurrence-free interval (19 months) were significantly better for PORT, with the complication rates being similar.

CONCLUSION

PORT had significantly better oncological outcomes compared to non-PORT. Multimodal management is worthwhile.