Predictors of gestational trophoblastic neoplasms chemotherapy outcomes at Kenyatta National Hospital:a retrospective cohort design study

Abstract

Background: Gestational trophoblastic neoplasms (GTN) span a spectrum of abnormal neoplastic trophoblastic proliferation that include choriocarcinoma, persistent Hydatidiform mole, invasive mole, placenta site trophoblastic tumour and epithelioid trophoblastic tumour. Except for the latter two, the tumours are highly curable with chemotherapy with remission rates reaching over 90%. The neoplasms are classified as low or high risk to single agent chemotherapy resistance depending on the WHO/FIGO scoring system. Treatment with single agent actinomycin-D or methotrexate for low risk and EMACO for high risk GTN are widely accepted standards of care. Justification: This study purposes to evaluate the GTN management and treatment outcomes at KNH in order to strengthen existing treatment guidelines and optimise patient treatment outcomes. No similar study has been carried out in comparable settings in sub-Saharan Africa. Broad Objective: To describe the management of gestational trophoblastic neoplasm at Kenyatta National Hospital and determine the predictors of chemotherapy treatment outcomes between 1st January 2010 and 31st December 2015. Methods: This is a retrospective cohort study with a calculated sample size of 156 using Kasiulevičius et al method with Fleiss continuity correction. All 158 patients treated at KNH between 1st January 2010 and 31st December 2015 whose records were accessed were analysed. Data was abstracted from the patients’ clinical records and entered in Epi Info version 7.1.2.0 for analysis. Univariate and bivariate analysis were used to calculate descriptive statistics and relative risks (RR) at 95% confidence interval. Chi square was used to determine statistical significance of various exposures on treatment outcomes. p values of less than 0.05 were considered statistically significant. Results: One hundred and fifty eight patient records were analysed that included 96 low risk and 62 high risk GTN patients. 42% of low risk GTN were treated with EMACO while as all but one of high risk patients were treated either with EMACO or EMA-EP. In addition, 14% and 7 % of patients required radiotherapy or surgery respectively. The overall remission rate was 65.2%. Methotrexate had a median of 5 courses and EMACO 4 courses to remission. Patients with WHO/FIGO score of 6 had

chemoresistance of 75% while on methotrexate while EMACO achieved 87.5% remission. Factors associated with treatment failure included use of single agent chemotherapy with RR 5.6 (95% CI 2.17 – 14.52 p<0.001), choriocarcinoma histopathology RR 2.9 (95% CI 1.37 – 6.30 p=0.015), term antecedent pregnancy RR 3.52 (95% CI 1.66 – 7.48 p=0.001), metastatic disease RR 1.98 (95% CI 1.56 – 2.48 P<0.001) and WHO/FIGO score of 6 or initial hCG >100,000 IU/L treated with methotrexate single agent RR 2.67 (95% CI 1.35 – 5.28 p=0.041). An hCG decline rate less than 10% between second and third courses of chemotherapy was predictive of treatment failure (p=0.03) with sensitivity of 60% and specificity of 72%. The median time lost from treatment initiation to outcome was significantly longer (p=0.015) for patients with treatment failure (45 days, IQR 16 - 52) compared to those with remission (22 days, IQR 12 - 37). Brain metastasis (100%) and choriocarcinoma histopathology RR 7.2 (95% CI 1.0 – 55.6) were invariably associated with death. Conclusion: The management of GTN at KNH does not strictly conform to WHO/FIGO guidelines. The treatment remission rate in the institution is significantly below that of comparable reference facilities. The identifiable predictors of treatment failure include high risk disease by WHO/FIGO score classification, choriocarcinoma histopathology, presence of metastasis, the rate of hCG decline between second and third course of chemotherapy less than 10%, and noncompliance with treatment protocol. Recommendations. The findings of this study validate an urgent need to standardize GTN care at KNH through a written guideline based on WHO/FIGO recommendations with regular monitoring of adherence to its protocols. Patients with histological diagnosis of choriocarcinoma and/or initial hCG >100,000 IU/L should be treated with EMACO irrespective of their WHO/FIGO score.