Gene expression profile in decisions for systemic therapy

<back

Decisions to administer chemotherapy in breast cancer are usually formulated on baseline risk assessment and clinical judgement is often used to weigh the degree of benefit from chemotherapy against risk and adverse effects.  Traditionally, this is based on standard pathological parameters such as tumour size, grade, lymphovascular invasion, hormone receptor status, HER-2 status and patient age.  Some of these rather gross variables form the basis of prognostic tools such as Adjuvant! Online.  Several components of these parameters such as tumour grade may lack reproducibility.  Gene expression profiling may provide a more robust means of evaluating factors that predict for prognostic outcomes.  However, conventional gene expression profiling methods require high quality RNA in tissue and therefore are dependant upon fresh or snap frozen tumour tissue.  More recent developments in quantitative real-time polymerase chain reaction (QRT-PCR) allow some tests to be done on paraffin embedded tumour tissue sections.

There are two commercially available gene expression profiles.  Oncotype DX is a gene expression profile based on 21 genes, 16 cancer related genes and 5 reference genes.  The assay was developed as a context-specific assay for patients with oestrogen receptor positive with node negative disease treated with tamoxifen.  This was clinically validated by patients who participated in the National Surgical Adjuvant and Bowel Project NSABP B-14.  The results are expressed as a recurrence score (RS) that ranges from 0-100: low risk RS less than 18; intermediate risk RS ≥ 18 and <31; high risk ≥ 31.  The 21 gene assay RS has been validated by several single institution studies.  Application of the RS to patients who participated in the NSABP B-20 trial show significant increases in disease recurrence free survival with chemotherapy (CMF) over tamoxifen at 10 years in the high risk group compared to very small benefits in the low and intermediate risk groups. The 21 gene assay is being evaluated in a prospective trial called PACT-1 / TAILORx. 

The MammaPrint test is based on a 70 gene signature assay.  In this platform, fresh tissue needs to be collected at the time of surgery and cannot be run on paraffin embedded tissue.  The MammaPrint 70 gene signature assay divides patients into good signature and poor signature groups and has been compared to other prognostic scores such as lymph node positive versus lymph node negative status, St. Gallen criteria and the National Institute of Health (NIH) high versus low risk groups.    A study by TRANSBIG to validate the 70 gene signature profile showed that classification of these patients into low and high risk groups is possible with high separation capability.  The clinical validation of this retrospective analysis is being investigated in a large prospective trial called MINDACT. 

Whilst it would be ideal to have a more robust prospective validation of these gene expression profiling systems, patient demand may well determine the clinical applicability and acceptance of this approach by consensus and specialist physician groups.  MammaPrint and Oncotype DX are already available for clinical use in breast cancer and the future of using these predictive scores in decision making for adjuvant systemic therapy is extremely encouraging.  Currently, the most useful clinical application of gene expression profiling systems is in patients with borderline recommendations for chemotherapy versus not that may tilt the balance for offering or not accepting chemotherapy.  Potential developments to incorporate additional genes of pathologic variables and biomarkers into the existing gene profile platforms will make it very likely that the use of these tests for prognostic and predictive outcomes to individualize breast cancer treatment will become more commonplace in the future.

<back | top^