MammaPrint was developed based on research performed at the Netherlands Cancer Institute and the Antoni van Leeuwenhoek Hospital in Amsterdam. The results of the study were published in well-known, peer-reviewed publications and showed that the 70-gene expression profile could predict the likelihood of distant metastasis occurring in lymph node negative primary breast cancer. These genes are involved in regulating various aspects of cellular function such as cell cycle, invasion, metastasis and angiogenesis.

click to enlarge

Figure 1: Clear classification of breast cancer tumor samples into high risk (poor signature) and low risk (good signature) of recurrence of the disease.

MammaPrint was validated in an independent external study at the Netherlands Cancer Institute and based on a consecutive series of 295 (151 node-negative and 144 node-positive) patients, <53 years (stage I and II).The data was published in the New England Journal of Medicine in 2002. In node-negative patients, the profile yielded a 97 % 10-year overall survival rate in the low risk group and a 50 % survival rate in the high risk group. The probabilities of these patients remaining free of distant metastases after 10 years were 87 % and 44 %, respectively. In node-positive patients, 10-year overall survival rates were 92 % (low risk) and 60 % (high risk) with probabilities of remaining free of distant metastases given as 83 % and 57 %, respectively. Multivariable hazard analysis showed that the profile was an independent predictive factor, and by far the strongest predictor of distant metastases (hazard ratio 4.6) when compared with clinical risk factors. The figure shows that, in a Kaplan-Meier analysis, the profile yielded better separation between high and low risk groups of node-negative patients than the St. Gallen criteria. The profile classified up to 40% of node negative patients into the low risk group. (i.e. those patients that could potentially be spared from chemotherapy) whereas the St. Gallen criteria classified only 15% of the patients into the low risk group.

click to enlarge

MammaPrint was further validated in an independent external study by the TRANSBIG consortium, which involved 302 patients from five European centers: Institute Roussy (Villejuif, France), Karolinska Institute (Stockholm, Sweden), Centre Rene Hugenin (Saint-Cloud, France), Guy’s Hospital (London, U.K.) and John Radcliff Hospital (Oxford, U.K.). The data was published in the Journal of the National Cancer Institute in September 2006. In the reference group as published, patients classified as LOW RISK had a 97% chance of survival after 10 years and 90% chance to be metastasis free after 10 years, without adjuvant treatment. The patients classified as HIGH RISK had a 65% chance of survival after 10 years and less than 71% chance to be metastasis free after 10 years, without adjuvant treatment.

click to enlarge

Figure 3: Kaplan-Meier curves by gene expression profiling risk groups, time to distant metastasis.
5-year: Low risk group: 0.95 (0.91-0.99) - High risk group: 0.78 (0.72-0.84)
10-year: Low risk group: 0.90 (0.85-0.96) - High risk group: 0.71 (0.65-0.78)

Lymph node status is considered to be one of the most powerful prognostic factors for operable breast cancer, with a direct relationship between the number of positive nodes and disease outcome.  However, approximately 30% of lymph node-positive patients will remain free of distant metastases without adjuvant chemotherapy. Identifying patients with lymph node-positive disease who are at low risk of recurrence could lead to changes in guidelines for adjuvant chemotherapy.

Most recently another second independent international consortium demonstrated the prognostic power of MammaPrint in patients with up to 3 positive lymph nodes. The data showed that MammaPrint can accurately identify a low risk group of lymph node-positive breast cancer patients with an excellent survival.

In the study, 241 tumor samples from breast cancer patients with 1-3 positive lymph nodes at 2 hospitals were selected and analyzed by the international TRANSBIG consortium. Patients that were classified as “high risk” for recurrence using MammaPrint had significantly worse 5- and 10-year survival rates than those that were identified as “low risk.”  The numbers indicate that MammaPrint has excellent prognostic power, even in patients having 1-3 positive lymph nodes.

click to enlarge

Figure 4: Kaplan-Meier curves: in breast cancer patients with 1-3 positive lymph nodes the MammaPrint expression profile can identify a low risk group with a 95% chance to be metastasis free after 10 years and a 10-year overall survival probability of 98%.

The MammaPrint assay was originally designed to classify younger breast cancer patients. In this study it could be demonstrated, that it can also accurately identify postmenopausal breast cancer patients at low risk or high risk of recurrence of the diesease.

In the study, breast tumors from 100 patients with lymph node-negative, invasive breast cancer (median age 62 years, but also including elderly patients over age 75) who were diagnosed and treated at the Massachusetts General Hospital in Boston, USA between 1985 and 1997 were subjected to MammaPrint gene expression analysis.  MammaPrint, having a digital readout, classified patients as being at either low or high risk for distant metastasis and the results were compared to actual outcome of disease in these patients.  The results of the study confirm that MammaPrint is also a very powerful tool to identify older breast cancer patients at low risk for metastatic disease.

click to enlarge

Figure 5: Kaplan-Meier curves of time to metastasis for node-negative patients from Massachusetts General Hospital, based on gene expression profiling (red: high risk, blue: low risk). The negative predictive value was 100% at 5 years in the MGH cohort.

The feasibility study “Mircoarray Prognostics in Breast Cancer (RASTER)” demonstrated that MammaPrint was successfully implemented in the diagnostic process of breast cancer patients in community hospitals in The Netherlands. MammaPrint assigned up to one-third of the patients to different risk categories as compared to currently used risk assessment tools. In the study, breast tumors from 585 patients with lymph node-negative, invasive breast cancer, were collected at 16 hospitals in The Netherlands. Results were available for 427 patients. Based on this, patients were classified as having either a poor or good prognosis. These results were then compared by the study coordinators to risk assessments of commonly used guidelines, including the Dutch Institute for Health Improvement (Dutch CBO) Guidelines, St. Gallen Guidelines, Nottingham Prognostic Index, and Adjuvant!Online. According to the results, adjuvant chemotherapy was advised less often when the patient’s prognosis was based on the DUTCH CBO Guidelines – which are considered the most restrictive – compared with assessments based on MammaPrint. However, for the other guidelines assessed, less adjuvant therapy would be given based on the MammaPrint prognosis. These results confirm previous findings and validations with the test.