Recurrence Online is capable to predict response to hormonal treatment [ER status], response to targeted therapy [HER2 status] and survival [recurrence score] for breast cancer patients using gene expression data obtained by Affymetrix microarrays. Quality control algorithms are implemented to exclude biases related to sample processing, hybridization and scanning.

What do I need to start using RecurrenceOnline?

You must have a .CEL file to use the system.

How was the RS calculator algorithm validated?

All together 2,472 unique microarrays with available clinical follow-up downloaded from GEO were used for validation. The datasets were generated using HGU133A or HGU133Aplus2 microarrays which possess overlapping probe set identifiers and are also used in by the www.recurrenceonline.com analysis. After classification of the samples the groups were compared using Kaplan-Meier survival plots. The RS calculator achieved a HR of 2.9 with a confidence interval of 2.5-3.3 and a p value <1E-16.

Custom validation

Anyone can validate the system on previously published microarrays. For this, you can download arrays from GEO (http://www.ncbi.nlm.nih.gov/gds). When searching for suitable datasets, use the keywords "breast cancer" and "gpl96" (for Affymetrix HGU133A microarrays) or "gpl570" (for Affymetrix HGU133Aplus2 microarrays).