3 The diagnostic tests

3 The diagnostic tests

3.1 The assessment compared 5 intervention tests with 1 comparator.

The interventions

EndoPredict (Myriad Genetics)

3.2 EndoPredict is a CE-marked assay that is designed to predict the likelihood of metastases developing within 10 years of an initial breast cancer diagnosis. The test is for pre and postmenopausal people with early breast cancer with oestrogen receptor (ER)‑positive, human epidermal growth factor 2 (HER2)-negative and lymph node (LN)-negative or LN-positive disease (up to 3 positive nodes).

3.3 EndoPredict measures the expression of 12 genes: 3 proliferation‑associated genes, 5 hormone receptor-associated genes, 3 reference (normalisation) genes and 1 control gene.

3.4 EndoPredict needs RNA extracted from a formalin-fixed, paraffin‑embedded (FFPE) breast cancer tissue sample. The test can be done in a local laboratory or the Myriad Genetics pathology laboratory in Germany. It takes approximately 2 days from receipt of the tissue sample to get the results from a local laboratory, although the turnaround time may be longer if samples are tested in full batches. The turnaround time is also longer if samples are sent to Germany.

3.5 The test involves a reverse transcription-quantitative polymerase chain reaction. Online evaluation software calculates an EP score and an EPclin score. An EP score of 0 to less than 5 indicates low risk of distant disease recurrence in the next 10 years. An EP score of 5 to 15 indicates high risk of distant disease recurrence in the next 10 years.

3.6 The EPclin score estimates the probability of metastases developing within 10 years (assuming 5 years of endocrine therapy). It is calculated by adding clinical data about tumour size and nodal status to the EP score. An EPclin score of less than 3.3 indicates low risk (less than 10%) of metastases in the next 10 years. An EPclin score of 3.3 or more indicates high risk of metastases in the next 10 years.

3.7 During consultation on the first diagnostics consultation document, NICE accepted an access proposal from the company in line with the Diagnostics Assessment Programme's interim addendum on access proposals. This provides a simple discount to the list price of EndoPredict, with the discount applied at the point of purchase or invoice. The level of the discount is commercial in confidence.

MammaPrint (Agendia)

3.8 MammaPrint is a CE-marked assay that is designed to assess the risk of distant recurrence within 5 and 10 years and whether a person would benefit from chemotherapy. The test is for pre and postmenopausal people with stage 1 or 2 breast cancer, with a tumour size of 5 cm or less, and LN-negative or LN-positive disease (up to 3 positive nodes). The test can be used irrespective of ER and HER2 status.

3.9 MammaPrint measures the expression of 70 genes, including genes associated with 7 different parts of the metastatic pathway: growth and proliferation, angiogenesis, local invasion, entering the circulation, survival in the circulation, entering organs from the circulation, and adaption to the microenvironment at a secondary site.

3.10 The MammaPrint test needs RNA extracted from an FFPE breast cancer tissue sample. The test is offered as an off-site service. In Europe, samples are analysed at the Agendia laboratory in the Netherlands. Results are available within 10 days of submitting the sample.

3.11 The test is based on diagnostic microarray. Software is used to calculate the MammaPrint result on a scale of −1 to +1. The score indicates the risk of developing distant metastases over the next 10 years without any adjuvant endocrine therapy or chemotherapy. A MammaPrint result of 0 or less indicates high risk of metastases in the next 10 years and a result of more than 0 indicates low risk (10% or less) of metastases in the next 10 years.

Oncotype DX Breast Recurrence Score (Genomic Health)

3.12 Oncotype DX Breast Recurrence Score (hereafter referred to as Oncotype DX) is designed to quantify the 10-year risk of distant recurrence and predict relative treatment effects for chemotherapy. The test also reports the underlying tumour biology: ER, progesterone receptor and HER2 status. The test is for pre and postmenopausal people with stage 1 or 2 breast cancer and ER‑positive, HER2-negative, LN-negative or LN-positive disease (up to 3 positive nodes). The assay does not have a CE mark because it is provided as a service by Genomic Health in an accredited laboratory in the US.

3.13 Oncotype DX quantifies the expression of 21 genes: 16 cancer-related genes correlated with distant recurrence-free survival, and 5 reference (normalisation) genes.

3.14 The Oncotype DX test needs RNA extracted from a FFPE breast cancer tissue sample. Samples are processed centrally at a Genomic Health laboratory in the US. Results are usually available 7 to 10 days after the sample is received.

3.15 The test is based on a reverse transcription-quantitative polymerase chain reaction. It gives a Recurrence Score result of between 0 and 100, which is used to quantify the 10-year risk of distant recurrence, assuming 5 years of endocrine therapy. A result below 18 indicates low risk of distant recurrence and claims to predict little to no effect of chemotherapy on patient outcomes. A result between 18 and 30 indicates intermediate risk of recurrence and claims to predict no substantial effect of chemotherapy on patient outcomes. A result of 31 or more indicates high risk of recurrence and claims to predict a large effect of chemotherapy on patient outcomes. NICE is aware that the cut-offs may change in light of the results from the TAILORx study.

3.16 The breast Recurrence Score result can be combined with clinical and pathological factors using the recurrence score-pathology-clinical (RSPC) calculator. However, this calculator has not been validated in a cohort independent of that used to derive Oncotype DX.

3.17 During consultation on the first diagnostics consultation document NICE accepted the company's commitment to maintain the current confidential discount, which is in line with the Diagnostics Assessment Programme's interim addendum on access proposals. This provides a simple discount to the list price of Oncotype DX, with the discount applied at the point of purchase or invoice. The level of the discount is commercial in confidence.

Prosigna (NanoString Technologies)

3.18 Prosigna is a CE-marked assay designed to provide information on breast cancer subtype and to predict distant recurrence-free survival at 10 years. The test is for postmenopausal people with early breast cancer that is ER-positive, HER2-negative and LN‑negative or LN-positive (up to 3 positive nodes).

3.19 Prosigna measures the expression of 50 genes used for intrinsic subtype classification, 8 housekeeping genes used for signal normalisation, 6 positive controls, and 8 negative controls.

3.20 The test needs RNA extracted from a FFPE breast tumour tissue sample. It is based on direct mRNA counting using fluorescent probes and an nCounter Digital Analyser.

3.21 Prosigna classifies the risk of distant recurrence within 10 years, assuming 5 years of endocrine therapy, based on the PAM50 gene signature, breast cancer subtype, tumour size, nodal status and proliferation score. The proliferation score is determined by evaluating multiple genes associated with the proliferation pathway. The test gives a score between 0 and 100. Based on this score and the nodal status, samples are classified into risk categories:

  • LN-negative: low risk (0 to 40), intermediate risk (41 to 60) or high risk (61 to 100).

  • LN-positive (up to 3 positive nodes): low risk (0 to 15), intermediate risk (16 to 40), or high risk (41 to 100).

3.22 During consultation on the first diagnostics consultation document, NICE accepted an access proposal from the company in line with the Diagnostics Assessment Programme's interim addendum on access proposals. This provides a simple discount to the list price of Prosigna, with the discount applied at the point of purchase or invoice. The level of the discount is commercial in confidence.

IHC4 and IHC4+C

3.23 The IHC4 test is a laboratory developed test that combines the results of 4 immunohistochemistry (IHC) measurements. The IHC4+C test combines the results of the 4 IHC4 tests with clinical and pathological features such as age, nodal status, tumour size, and grade. Both versions are designed to quantify the 10-year risk of distant disease recurrence, assuming 5 years of endocrine therapy. The test is for postmenopausal people with early breast cancer that is ER-positive and LN-negative or LN-positive (up to 3 positive nodes).

3.24 The IHC4+C test needs an FFPE breast tumour tissue sample. The 4 immunohistochemistry tests are: ER, progesterone receptor (PR), HER2 and the proliferation marker Ki67. ER and HER2 markers are commonly measured in NHS laboratories, but PR and Ki67 markers are not.

3.25 The IHC4+C test is in clinical use at 1 NHS centre (the Royal Marsden NHS Foundation Trust), which carries out the test with an average turnaround time of 1 week. The test could be run in local NHS laboratories providing that training and quality assurance programmes for the individual assays were in place.

3.26 The IHC4+C test uses a published algorithm to calculate a risk score for distant recurrence based on the results of the 4 assays and clinical factors. A calculator is available for use on request. A score of less than 10% is categorised as low risk for distant recurrence at 10 years. A score of more than 10% but less than 20% is intermediate risk, and a score of 20% or more is high risk for distant recurrence at 10 years.

The comparator

3.27 The comparator is decision making for adjuvant chemotherapy prescribing, based on clinical and pathological features or the results of tools used to assess risk without the tumour profiling tests. Features may include the stage and grade of the disease, nodal status, ER or PR status, HER2 status and any previous treatment (for example, neoadjuvant therapy). Risk assessment tools include PREDICT, the Nottingham Prognostic Index (NPI) and Adjuvant! Online.

3.28 These risk assessment tools can be used to define the level of clinical risk. For example, in LN-negative disease a NPI of 3.4 or less is classed as low risk, and a NPI of more than 3.4 and up to or equal to 5.4 is classed as intermediate risk. If using the PREDICT tool, an absolute 10-year survival benefit from chemotherapy of less than 3% is classed as low risk; between 3% and 5% is classed as intermediate risk; and more than 5% is classed as high risk.

  • National Institute for Health and Care Excellence (NICE)