An assessment of survival outcomes among ovarian cancer patients at the National and Referral Hospital in Kenya

Abstract Background Ovarian cancer has been shown to have poor survival outcomes attributed to late presentation. In Kenya, information on the survival outcomes of ovarian cancer patients is scarce. Therefore, the objective of this study was to examine the survival outcomes among patients with ovarian cancer treated at Kenyatta National Hospital (KNH). Aims A hospital‐based retrospective cohort study was performed at KNH to examine the survival outcomes of 112 ovarian cancer patients. The study employed a structured data abstraction tool to acquire patients' relevant socio‐demographic and clinical characteristics from the patient's medical records. The data obtained were analyzed using SPSS version 29.0 statistical software. Kaplan–Meier and Cox regression analyses were used to determine the survival outcome and predictors of mortality among ovarian cancer patients, respectively. Methods and results The mean age of the patients in this study was 51.28 ± 14.24 years. Most patients (59.8%) had evidence of distant metastasis during the follow‐up period. One‐third (33%) of patients were deceased. The mean‐cancer‐specific survival time among the study participants was 40.0 ± 3.0 months. The 5‐year survival rate was 44%, with most patients experiencing disease progression during the last follow‐up. Combination therapy (p < .001) was the only statistically significant predictor of mortality in ovarian cancer patients. Conclusion The study found that the 5‐year survival rate among ovarian cancer patients was poor, with most patients experiencing disease progression during the last follow‐up period.


| INTRODUCTION
Among females, ovarian cancer ranks as the third most prevalent gynecological malignancy, following cervical and uterine cancer. 1 Ovarian cancer constituted 1.6% of 18.1 million new cancer diagnoses and 1.9% of the 9.6 million global cancer-related fatalities. 2 In contrast to breast cancer, ovarian cancer has a lower incidence rate, yet it carries a higher mortality rate.The mortality rate due to ovarian cancer is estimated to significantly increase by the year 2040. 1 In the African continent, approximately 25 000 new cases and 17 000 fatalities of ovarian cancer were recorded in 2020, 1 with a mortality-to-incidence ratio of 0.7, making ovarian cancer among the top ten lethal cancers. 3A recent study in Sudan showed that most ovarian cancer patients presented with an advanced stage of the disease, resulting in shorter survival times. 4Nonetheless, another study showed that early detection does not improve overall survival. 5In contrast, other studies have reported that ovarian cancer survival can be improved by early detection. 6,7e emergence of newer treatments in recent years, including intraperitoneal chemotherapy and treatment with antiangiogenic and targeted chemotherapies, has improved survival rates. 8Despite this, the 5-year survival rates of patients with ovarian cancer remain low. 4,9Therefore, achieving the desired goal of improved survival rates remains challenging.
Moreover, there is a paucity of adequate data on the survival outcomes of ovarian cancer patients in Kenya.Thus, the primary objective of this study was to evaluate the survival outcomes experienced by patients with ovarian cancer receiving treatment at Kenyatta National Hospital (KNH).

| Study design
A hospital-based retrospective cohort study was employed to evaluate the survival outcomes and the associated factors among ovarian cancer patients who received treatment at the KNH from January 1st, 2017 until December 31st, 2021.

| Study setting and period
This study was conducted for 6 months (October 1st, 2022 to March 31st, 2023) at KNH.The hospital was founded in 1901 and is located along Hospital Road in Nairobi, Kenya.It is one of the largest referral and teaching hospitals in Kenya.

| Study population
The study population included all eligible adult ovarian cancer patients who had been treated at the KNH from January 1st, 2017 until December 31st, 2021.Patients who had a complete interval imaging results to assess the response of the tumor.

| Exclusion criteria
Patients with incomplete medical records on their treatment regimen, cancer diagnosis, and cancer stage were excluded from the study.
The details of the eligibility criteria are stated in Figure 1.

| Sample size determination
The health records department at KNH identified ovarian cancer cases treated at their facility through an electronic search by use of the International Classification of Diseases, 10th Revision (ICD-10) Diagnosis Code C56, which is specific for malignant neoplasm of the ovary.The files at the facility registered under the code were filtered by the system to include ovarian cancer cases according to the eligibility criteria and 712 cases were identified.The identified files were manually extracted from the file's storage, of which only 250 files could be traced, and of the total, only 112 files had complete relevant information required for the study.In this study, therefore, 112 eligible patients comprised the study's sample size (Figure 1).

| Data collection tools and procedures
A data abstraction tool was used to collect data from the medical records of patients.The data abstraction tool contained essential information, such as socio-demographic factors of the patients, clinical characteristics, and survival outcome measuring parameters.1][12][13] A pretest was conducted on 5% of the total sample size.Before employing the data collection tool in the main study, all necessary adjustments were made.The Health Information Department of KNH was consulted to obtain the patient's medical records.The research assistant and principal investigators conducted a manual extraction of data from the patient's medical records.Subsequently, a comprehensive review was performed among them to ensure data completeness and accuracy, with any discrepancies meticulously examined and resolved through adjudication.Relevant information was assessed, including the patient's socio-demographics, treatment history, time of death, length of survival, and response to therapy.Mortality and disease progression were assessed as outcomes.The data obtained were then reviewed using appropriate data analysis tools.

| Data analysis
The data collected were entered and analyzed using SPSS version 29.0 software.Socio-demographic factors, such as patient age, were described using mean and standard deviation.On the other hand, marital status, level of education, and employment status were presented using frequency and percent.Clinical characteristics, treatment regimens and treatment outcomes including tumor size after treatment, patient status, and the presence of distant metastasis among others, were also presented through frequency and percentage tables.The mean cancer-specific survival was calculated from the months between the initial diagnosis and death or last follow-up.Metastasisfree survival was determined from the months between the initial diagnosis and the first occurrence of radiologic metastasis.Cancer-specific survival after metastasis was computed from the months between the first occurrence of radiologic metastasis and death or last follow-up.
The data were computed by calculating an average of the months.
Survival outcomes for ovarian cancer patients were assessed through Kaplan-Meier analysis.The 5-year survival rate analysis focused on ovarian cancer patients who survived for at least 5 years from the initial diagnosis.Cox regression analysis, was conducted to estimate predictors of mortality.In bivariate Cox regression analyses, independent variables (e.g., age, cancer stage, histological type, comorbidity, distant metastasis, and treatment modalities) were compared with the outcome variable (censored or mortality).In multivariate Cox regression analyses, the independent variables (stage of cancer, distant metastasis, and type of treatment regimens) were collectively assessed against the outcome variable to identify predictors of mortality.Variables with a p-value above .10 in the bivariate analyses (age, histological type, and comorbidity) were excluded from multivariable analyses.

| Socio-demographic characteristics
The mean age of the patients in this study was 51.28 ± 14.24 years.
Regarding the employment status, most of them were unemployed F I G U R E 1 Inclusion and exclusion criteria among ovarian cancer patients.

| Clinical characteristics
Moreover, the median follow-up time of the patients in this study was 13.5 months and the maximum and minimum follow-up time was 63 months and 1 month, respectively.The predominant histological type of ovarian cancer, epithelial ovarian cancer, was observed in 82.1% of the patients, followed by sex cord-stromal ovarian cancer (12.5%) and germ cell cancer (5.4%).Most patients (50.9%) presented with stage IV disease at diagnosis.The most prevalent co-morbidity was hypertension, accounting for 29.5% of all patients (Table 2).
The standard treatment of the majority of the patients (65.2%) was surgery and chemotherapy, followed by surgery (28.6%).Most patients (36.6%) underwent six cycles of chemotherapy to manage their malignancy (Table 3).On the trend of diagnosis, it was observed that most patients were diagnosed in year 2019 (220), out of the total 712 cases over the study period (Figure 2).

| Treatment outcomes
Most patients (59.8%) had evidence of distant metastasis during the follow-up period, whereas 40.2% had no evidence of distant metastasis.Two-thirds (67.0%) of patients were censored, and 33% died.For most patients (42.9%), the CA-125 levels significantly increased during the follow-up period (Table 4).The meancancer-specific survival time among the study participants was 40.0 ± 3.0 months.On the other hand, the metastasis-free survival and cancer-specific survival after metastasis were 15.5 ± 4.3 months and 16.7 ± 2.3 months, respectively (Table 5).The present study showed that the 5-year survival rate of ovarian cancer patients was 44%.

| Predictors of mortality
The Cox regression was used to perform bivariate and multivariate analyses to investigate the association between the independent variables and the risk of death.The variables used were age, cancer stage, histological type, co-morbidity, distant metastasis, and treatment regimens.Combination therapy was the only statistically significant predictor of mortality in ovarian cancer patients (Table 6).Nonetheless, in the log-rank test, there was a significant difference in survival rate  among different stages of cancer and treatment regimens (Figures 3   and 4).

| DISCUSSION
This retrospective study examined survival outcomes and factors linked to ovarian cancer patients at KNH.In terms of the occurrence pattern, the results in this study indicated that most cases of ovarian cancer were reported in patients below 60 years (65%) with a mean age of 51.28 ± 14.24 years.In Taiwan, the mean age at diagnosis was 52.8 ± 11.2 years. 14Moreover, previous research has found that the median age at diagnosis ranges from 50 to 79 years. 15A recent study reported that most ovarian cancer cases were above 50 years. 16This is consistent with the findings in Nigeria, where ovarian cancer was most prevalent around the fifth decade of life. 17Although several studies have reported that ovarian cancer diagnosed at a younger age has a better prognosis, [18][19][20] other researchers have argued that age is not a standalone prognostic factor. 21ithelial, stromal, or germ-cell tumors account for most benign and malignant ovarian cancers.Various studies have reported that in developed countries, up to 90% of all malignant ovarian cancers have an epithelial origin, 5%-6% of the tumors are sex-cord stromal tumors and 2%-3% originate from the germ cells. 18,22Similarly, this study found that epithelial ovarian cancer   Previous studies have reported that the stage at detection has a considerable impact on the survival of ovarian cancer patients. 14,23st ovarian cancer patients present with an advanced stage of the disease due to the insidious onset of nonspecific symptoms and a lack of timely and proper screening. 6However, a study conducted in Taiwan reported that a relatively high percentage of the patients  (44.4%) had an early-stage disease, which could be attributed to the ease of access to undergo early screening. 23In the present study, the majority of the patients (50.9%) presented with stage IV disease at diagnosis, which can be ascribed to poor health-seeking habits as a result of the study participants' low socioeconomic status and level of education. 24third (37.5%) of the total study population presented with comorbidities, with hypertension being the most common type (29.5%), followed by diabetes (8.0%).In general, there is an established correlation between hypertension and an elevated risk of developing cancer.
A recent study suggested that hypertension causes dysregulation of apoptosis, therefore, increasing cancer risk. 25Additionally, another study hypothesized that elevated angiotensin II levels in hypertensive patients can stimulate the synthesis of vascular endothelial growth factor, which augments cancer-related angiogenesis. 26A cohort study on cancer patients reported hypertension as the most common comorbidity. 27A recent study conducted in Saudi Arabia reported diabetes mellitus and hypertension as the most common co-morbidities affecting ovarian cancer patients since each of them accounted for 39.5% of the total study population. 28Another study reported hypertension (11%-26%), cardiovascular disease (4.5%-12%), and diabetes (2.5%-8.3%)as the most prevalent comorbidities in ovarian cancer patients. 29e recommended standard treatment guideline for ovarian cancer involves surgery followed by platinum-based combination chemotherapy. 30Similarly, patients who received adjuvant combination chemotherapy of a taxane-based regimen were reported to have considerably improved overall survival than those who received nontaxane-based chemotherapy regardless of the stage of cancer. 14other study also reported similar results in ovarian cancer patients who received the taxane-based combination chemotherapy had significantly improved post-recurrence survival. 31 The disease progression and treatment efficacy in ovarian cancer patients are monitored using changes in CA-125 levels. 32Various studies have found a relationship between increasing CA-125 levels posttreatment with an increased risk of ovarian cancer recurrence. 33,34Consequently, the increasing CA-125 levels in the first months posttreatment may indicate resistance to platinum-based chemotherapy. 35In the present study, for most patients (42.9%), the CA-125 levels significantly increased during the follow-up period, with 36.6% having the CA-125 levels significantly decreased and 20.5% experiencing no change in the CA-125 levels posttreatment.
In the present study, the 5-year survival rate of ovarian cancer patients was 44%.This finding is similar to a study conducted in Canada, which reported a 5-year survival rate of 44%. 18Another European cohort study reported 5-year survival rates ranging from 26% to 51%. 23A study conducted in Sudan reported that the 5-year cumulative survival rate was 38% for all histological types of ovarian cancer. 4The relatively low survival rates reported among ovarian cancer patients could be due to most of them presenting with an advanced stage of the disease, as was observed in our setting, which could be attributed to the delayed onset of symptoms and the lack of factor influencing survival outcomes was combination therapy (surgery and chemotherapy only) among ovarian cancer patients.

| Limitations of the study
The study was retrospective in design; therefore, it relied on the effective documentation of medical records, which was lacking, as evidenced by a high percentage of missing data.

| CONCLUSION
The 5-year survival rate among ovarian cancer patients was poor, with most patients experiencing disease progression during the last followup period.

3 | ELIGIBILITY CRITERIA 3 . 1 |
Inclusion criteria Medical records of patients aged 18 years and above with a confirmed diagnosis of ovarian cancer treated from January 1st, 2017 until December 31st, 2021.Patients who had comprehensive medical records detailing their treatment regimen, cancer stage, and details on their diagnosis.
For the tumor size after treatment, the baseline for tumor size measurement was established using computerized tomography (CT) scan results.The increase or decrease in tumor size was assessed by comparing posttreatment CT scans with the baseline scans.

T A B L E 1
Socio-demographic characteristics of the study participants (N = 112).

T A B L E 3
Treatment regimen of the study participants (N = 112).

2 6 T A B L E 5
Trend of diagnosis among ovarian cancer patients.T A B L E 4 Treatment outcomes of ovarian cancer patients (N = 112).Survival parameters among the study participants.the most common histological type, accounting for 82.1% of all ovarian cancer cases.
Adjusted for age, stage of cancer, histological type of cancer, comorbidity, distant metastasis, and type of treatment regimen.Combination therapy (surgery and chemotherapy).Abbreviations: AHR, Adjusted hazard ratio; CHR, Crude hazard ratio.*Statistically significant with p-value <.05.F I G U R E 3 Kaplan-Meier survival curve of ovarian cancer patients based on tumor stage.
In our setting, twothirds (65.2%) of the patient population were treated with surgery and chemotherapy.The most common chemotherapy regimen administered was adjuvant cisplatin and paclitaxel.Moreover, most patients received taxane-based combination chemotherapy composed of a platinum drug such as cisplatin or carboplatin with either paclitaxel or docetaxel, hence the significant reduction of tumor size after treatment observed in close to half (49.1%) of the study population.
health-seeking behavior among ovarian cancer patients.Additionally, with the advanced tumor stage, the tumor has metastasized to other organs in the body, and the likelihood of recurrence of the tumor is high despite initial treatment.Several variables that have been associated with predicting survival in ovarian cancer include the stage at detection, histologic subtype, age, tumor grade, postoperative residual tumor, response to chemotherapy, co-morbidities, and pretreatment serum concentrations of CA-125. 9,36Age, histological type, stage of cancer, distant metastasis, co-morbidities, and treatment regimen were the factors considered in our setting.Of these variables, the only significant F I G U R E 4 Kaplan-Meier survival curve of ovarian cancer patients based on treatment regimen.
Predictors of mortality among ovarian cancer patients.
T A B L E 6