Corresponding author: M J Kotze (maritha@sun.ac.za)Background. Clinical and pathological parameters may overestimate the need for chemotherapy in patients with early-stage breast cancer. More accurate determination of the risk of distant recurrence is now possible with use of genetic tests, such as the 70-gene MammaPrint profile. Objectives. A health technology assessment performed by a medical insurer in 2009 introduced a set of test eligibility criteria -the MammaPrint Pre-screen Algorithm (MPA) -applied in this study to determine the clinical usefulness of a pathology-supported genetic testing strategy, aimed at the reduction of healthcare costs.Methods. An implementation study was designed to take advantage of the fact that the 70-gene profile excludes analysis of hormone receptor and human epidermal growth factor receptor 2 (HER2) status, which form part of the MPA based partly on immunohistochemistry routinely performed in all breast cancer patients. The study population consisted of 104 South African women with early-stage breast carcinoma referred for MammaPrint. For the MammaPrint test, RNA was extracted from 60 fresh tumours (in 58 patients) and 46 formalin-fixed, paraffin-embedded (FFPE) tissue samples. Results. When applying the MPA for selection of patients eligible for MammaPrint testing, 95 of the 104 patients qualified. In this subgroup 62% (59/95) were classified as low risk. Similar distribution patterns for risk classification were obtained for RNA extracted from fresh tumours v. FFPE tissue samples. Conclusions. The 70-gene profile classifies approximately 40% of early-stage breast cancer patients as low-risk compared with 15% using conventional criteria. In comparison, more than 60% were shown to be low risk with use of the MPA validated in this study as an appropriate strategy to prevent chemotherapy overtreatment in patients with early-stage breast cancer.

Molecular profiling and functional assessment of signalling pathways of advanced solid tumours are becoming increasingly available. However, their clinical utility in guiding patients’ treatment remains unknown. Here, we assessed whether molecular profiling helps physicians in therapeutic decision making by analysing the molecular profiles of 1057 advanced cancer patient samples after failing at least one standard of care treatment using a combination of next-generation sequencing (NGS), immunohistochemistry (IHC) and other specific tests. The resulting information was interpreted and personalized treatments for each patient were suggested. Our data showed that NGS alone provided the oncologist with useful information in 10–50% of cases (depending on cancer type), whereas the addition of IHC/other tests increased extensively the usefulness of the information provided. Using internet surveys, we investigated how therapy recommendations influenced treatment choice of the oncologist. For patients who were still alive after the provision of the molecular information (76.8%), 60.4% of their oncologists followed report recommendations. Most treatment decisions (93.4%) were made based on the combination of NGS and IHC/other tests, and an approved drug- rather than clinical trial enrolment- was the main treatment choice. Most common reasons given by physicians to explain the non-adherence to recommendations were drug availability and cost, which remain barriers to personalised precision medicine. Finally, we observed that 27% of patients treated with the suggested therapies had an overall survival > 12 months. Our study demonstrates that the combination of NGS and IHC/other tests provides the most useful information in aiding treatment decisions by oncologists in routine clinical practice.

Approximately 25 % of clinically important drugs and numerous environmental carcinogens are metabolised by CYP2D6. Variation in the CYP2D6 gene and concomitant use of tamoxifen (TAM) with certain antidepressants may increase recurrence risk in breast cancer patients due to reduced enzyme activity. In this study we determined the appropriateness of adding CYP2D6 genotyping to the breast cancer genetic testing options already available in South Africa, which include BRCA mutation screening and transcriptional profiling to assess estrogen receptor (ER) status. A total of 114 South African breast cancer patients, including 52 Caucasian and 62 Coloured (Mixed ancestry), and 63 Caucasian control individuals were genotyped for the most common inactivating allele (CYP2D6*4, rs3892097) previously identified in the CYP2D6 gene. In the initial validation data set consisting of 25 Caucasian and 62 Coloured patients, the CYP2D6*4 allele frequency was significantly higher in Caucasian compared to Coloured patients (24 % vs. 3 %, p < 0.001), similar to previous findings in the general South African population. Extended CYP2D6 genotyping was subsequently performed in an implementation data set of 27 Caucasian breast cancer patients, to determine the prevalence of depression and use of antidepressants in a clinical setting. A medical history of depression and/or use of antidepressants was reported in 37 % (10/27) of these breast cancer patients genotyped for CYP2D6*4. This translational research study has led to increased awareness among clinicians of the potential benefits of CYP2D6 genotyping to facilitate prevention of cumulative risk in a high-risk genetic subgroup of breast cancer patients considered for concomitant treatment of TAM and antidepressants that may reduce enzyme function.

No abstract

No abstract

Defects in the BRCA tumor suppressor genes contribute significantly to the development of breast cancer in South Africa. Additive genetic effects and lifestyle risk factors underlie variable expression patterns in affected families, while also increasing breast cancer risk in the general population. Modifiable environmental factors could determine whether individuals with genetic risk factors develop cancer or fail to respond to treatment. Consequently, there is an increasing demand for genetic testing not only in patients with familial breast cancer, but also in healthy women and patients with sporadic cancer. Today, genetic testing in breast health extends from single-gene diagnostic tests to multi-gene treatmentbased tests. For women in whom established breast cancer risk reduction approaches such as surgery and pharmaceuticals are not acceptable or inappropriate, treatment is based on molecular-genetic targets for individualised medical, nutrition and lifestyle intervention. Genetic testing as part of a holistic approach to breast health management translates into individual needs-based risk reduction interventions.

No abstract

To compare the cost-effectiveness of various cervical cancer (CxCa) screening algorithms to primary screening with the cobas HPV Test, which identifies HPV genotypes 16/18 individually while simultaneously detecting the other high-risk HPV types. METHODS: A cohort Markov model was developed to compare four CxCa screening strategies: (S1) cytology with reflex HPV, (S2) cytology and HPV co-testing, (S3) HPV with reflex cytology, and (S4) cobas HPV with genotyping and reflex cytology. Screening began at age 30 with a routine screening interval of every 3 years, and was modeled over a time horizon of 40 years. Performance of the overall screening strategies, ie, sensitivity and specificity for CIN 2/3, was derived from the ATHENA (Addressing THE Need for Advanced HPV Diagnostics) trial. Trial baseline data were used for the base case, and 1-year follow-up outcomes were estimated for the alternative scenario, assuming all persistent disease is detected in the subsequent visit. The direct costs for screening and treatment of CxCa were estimated from a US payer perspective in 2010 US dollars. Costs and quality-adjusted life years (QALYs) were discounted at 3% annually. One-way sensitivity analyses were conducted. RESULTS: Using a $50,000/QALY threshold, baseline screening with S4 dominated S2 and S3 by reducing overall cost, annual cancer incidence, and improving QALYs; and was cost-effective compared to S1. In the 1-year follow-up scenario, S4 was cost-effective compared to all other strategies. Detection of HPV 16/18 with S4 resulted in earlier diagnosis of clinically relevant CIN 2/3 at the initial visit as well as more efficient use of screening tests during follow-up. Sensitivity analyses showed that the model results were most influenced by the costs of tests used. CONCLUSIONS: Incorporating the cobas HPV test with HPV 16/18 genotyping was cost-effective compared to various CxCa screening strategies, and resulted in improved protection against CxCa.