INTRODUCTION: Breast cancer makes up 25% of all cancers diagnosed worldwide. Despite an increasing yearly incidence, there has been a significant decrease in mortality owing to early diagnosis and advances in treatment. Anthracycline-based chemotherapy is a relatively low cost yet highly effective anti-cancer treatment, increasing survival from 30% to >80%, presently. However, treatment efficacy is marred by the increased risk of anthracycline-induced cardiotoxicity (ACT) - estimated at 10-26%. Internationally, there has been evidence of ACT having a genetic basis. Currently in South Africa, there is little information on ACT in cancer patients and survivors, and no information on the genetic basis of this phenomenon. Our recruitment sites in Cape Town - Groote Schuur Hospital (GSH) and Tygerberg Hospital (TBH), routinely treat hundreds of patients, notably with breast cancer, with anthracycline-based therapy every year, and provided the environment to assess ACT, as well as genetic factors which may influence this adverse drug reaction. Left ventricular ejection fraction (LVEF) acts as a surrogate measure of cardiac function in the public health-care setting. OBJECTIVES: To provide insight into the clinical management of breast cancer patients on anthracycline-based treatment with a focus on the prevalence of ACT. To provide an index of genetic susceptibility to ACT and potentially contribute to a personalized medicine approach for a genetically diverse population. METHODOLOGY: In the retrospective part of the study, the clinical records of cancer patients treated with anthracyclines from 2011- 2016 at the Oncology Clinic at GSH were analysed. Clinical co-morbidities such as hypertension, diabetes, pre-existing cardiac disease and smoking as well as type and dose of anthracyclines, cardiac function and patient status were assessed. In the prospective study, breast cancer patients treated with anthracyclines, with a pre and post-treatment LVEF measure were recruited at GSH and TBH from 2013 to 2016. Patients were consented for access to both clinical information and biological material. Demographics, clinical risk factors and chemotherapeutic regimen data were analysed. LVEF, biomarkers and clinical status were also assessed in terms of reflecting ACT. In some instances certain clinical information was not available (i.e. LVEF) and out of necessity, a statistical correlation model or classifier was created in order to use available clinical data to derive missing clinical measures. Patients' DNA were analysed for seven genetic variants in the following six genes ABCC1 (rs246221); ABCC2 (rs17222723; rs8187710); HNMT (rs17583889); NCF4 (rs1883112): RAC2 (rs13058338) and RARG (rs2229774), and tested for correlation with clinical status and cardiac injury. Finally, a corollary study was conducted on a subset of patients in an attempt to determine whether cardiac biomarkers may be more sensitive measures of cardiotoxicity. RESULTS & DISCUSSION: In the retrospective cohort (n=402) 19.7% of patients showed diminished cardiac function. Logistic regression showed that the following predictors: type of first line chemotherapy, and total dose significantly contributed to the ACT phenotype as measured by change in LVEF. In the prospective patients (n=272), 14% were affected with ACT, with an increased likelihood of cardiotoxicity in the Indigenous African population. Logistic regression showed that both total anthracycline dose and change in LVEF were predictive of ACT. In the association study of prospective patients, only the RARG rs2229774 variant was significantly associated with patient ACT status (p=0.049, Chi-Square Test). Forty-two patients were assessed for the β-Natriuretic Peptide (BNP) biomarker and showed limited utility in correlating clinical status and/or LVEF decrease in all patients except Indigenous Africans indicating potential increased susceptibility of population group to ACT. LVEF was found to be unreliable as significant LVEF decreases did not always correlate with cardiac impairment and vice-versa. Changes in routine clinical patient management and overburdening of the nuclear medicine department also translated to only one LVEF measure being obtained in some instances. The statistically derived classifier for missing indicators of heart function was useful, but will require refinement. CONCLUSIONS AND RECOMMENDATIONS: Despite the inability of genotype as a predictor of ACT in this study, the increased susceptibility in the Indigenous African population to ACT as well as increased BNP levels after chemotherapy requires a closer look. The interrogation of lndigenous African patient genomes for novel variants of susceptibility to ACT are recommended; this requires building up of a substantial cohort from this population group, which would likely require collaboration with health care institutions in one of the other provinces of South Africa e.g. Eastern Cape, KwaZulu-Natal and/or Gauteng. Both this study and literature recommend the need for clinical trials for new and existing drugs on local African populations for both safety and efficacy. Furthermore, the BNP biomarker may be better suited to the prediction of irreversible cardiac damage rather than early cardiotoxicity. Troponin, released in response to cardiomyocyte death, may be a more sensitive biomarker in predicting ACT. Similarly, the inherent variability and lack of sensitivity of LVEF as a measure of cardiac function warrants the consideration of alternatives such as echocardiography or tissue-doppler imaging. Findings derived from this study indicate the need for refined patient management of ACT in a South African population to potentially allow for treatment with minimised risk and event-free breast cancer survival.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/27839 |
| Date | January 2018 |
| Creators | Naidoo, Horacia |
| Contributors | Ramesar, Rajkumar, Simonds, Hannah M |
| Publisher | University of Cape Town, Faculty of Health Sciences, Division of Human Genetics |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Doctoral Thesis, Doctoral, PhD |
| Format | application/pdf |
Page generated in 0.0024 seconds