Key points from the evidence

3. The RxPONDER randomised controlled trial (RCT) was published after the NICE guidance. The RxPONDER trial evaluated the use of Oncotype DX^® in predicting the benefit of adjuvant chemotherapy in women with lymph node positive (LN+) disease. In addition, updated results (median follow up of 8.7 years) of an RCT, which evaluated the clinical utility of MammaPrint^® in women with early-stage breast cancer (the MINDACT trial) were published after the NICE guidance.

4. The collective evidence suggests that all four tumour profiling tests provide prognostic
information on a patient’s future risk of cancer recurrence and/or survival. The tests add prognostic value over other prognostic clinical and pathological information available to clinicians and patients. The evidence is weaker and more variable in patients with LN+ disease, compared with those who have LN- disease.

5. Predictive ability relates to the ability of the tests to identify those patients who are most likely to benefit from chemotherapy. Three of the tests are indicated for predictive use (Oncotype DX^®, MammaPrint^® and EndoPredict^®), but there is considerable uncertainty within the evidence that the tests are predictive of improved outcomes with chemotherapy.

6. A qualitative study published after the NICE guidance explored ways in which women in the UK interpret and discuss the Oncotype DX^® test. In addition, a patient organisation submission was received from Breast Cancer Now.

7. Patients generally view tumour profiling tests positively, as they are perceived as
providing ‘personalised’ information, which is seen as more reliable and informative than risk scores calculated from online algorithms. Low- and high-risk scores from the tumour profiling tests are generally viewed as providing more clarity around the decision to have or forgo chemotherapy, compared with intermediate-risk scores. Waiting for test results prolongs patient anxiety, and so the results of tumour profiling tests should be made available to patients in a timely manner.

8. SHTG’s updated literature review of the cost-effectiveness evidence highlighted six economic evaluations that were published after the NICE guidance, though these were more limited than the analyses presented by NICE at addressing the decision problem faced by NHSScotland.

9. The manufacturers of the four tests were invited to submit economic evaluations, and were received from the manufacturers of MammaPrint^®, Oncotype DX^® and Prosigna^®. The results of economic analyses submitted by test manufacturers strengthened the conclusions of the NICE recommendation from 2018, that tumour profiling tests (Oncotype DX^®, Prosigna^® and EndoPredict^®) are likely cost effective in patients with LN- disease who have an intermediate risk of recurrence using a validated algorithm. The MammaPrint^® test is also likely cost effective in this population. Cost effectiveness in this subgroup is driven by the avoidance of unnecessary chemotherapy and reducing distant recurrence.

10. Tumour profiling tests are unlikely to be cost effective for patients with LN- disease and a low risk of recurrence based on a validated algorithm as this subgroup have low rates of chemotherapy in current clinical practice without a test.

11. While three of the tumour profiling tests under consideration (EndoPredict^®, MammaPrint^® and Oncotype DX^®) may be cost effective in patients with LN+ (1-3 nodes) disease, these results should be treated with caution because of uncertainty in the clinical evidence base for tumour profiling tests in this subgroup.