Researchers have developed a multistate model to identify 11 integrative subtypes of breast cancer with different risks of recurrence, according to the findings of a study published online in Nature.

Although it is known that women with definitively treated early-stage breast cancer, particularly estrogen receptor (ER)-positive cancer, can experience disease recurrence up to 20 years following initial diagnosis, long-term follow-up data of molecularly characterized cohorts of breast cancer are lacking. Nevertheless, early identification of women at high risk for breast cancer recurrence remains important.

The multistate model described in the paper accounts for different disease states, time scales, competing risks of mortality, and baseline characteristics across different molecular subgroups to predict individual risk of breast cancer recurrence. This model was fitted to a dataset from 3240 patients diagnosed with breast cancer between 1977 and 2005, including 1980 patients with corresponding molecular data. The median clinical follow-up of this cohort was 14 years. Other smaller datasets were used for external validation of the model.

Specifically, 4 integrative subtypes, accounting for about 26% of ER-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer, were identified as having a high risk of late recurrence. Some of the molecular features of these subtypes included an enrichment in copy number alterations in likely driver genes, such as CCND1, FGF3, EMSY, PAK1, RSF1, ZNF703, FGFR1, RPS6KB1, and MYC. Interestingly, several of these genes are therapeutically targetable.

In addition, specific subgroups of triple-negative breast cancer were also identified as being associated with either an increased risk of recurrence many years after initial treatment, or a very low risk of distant recurrence in ER-positive/HER2-negative patients who were relapse-free at 5 years.

“Integrative subtyping may help to determine whether women who are relapse free five years after diagnosis might benefit from extended endocrine therapy or other interventions to improve late outcomes. Critically, the four late-recurring ER-positive subgroups are enriched for genomic-copy-number driver alterations that can be therapeutically targeted, paving the way for new treatment strategies for these high-risk patient populations,” the study authors concluded.

Reference

1. Rueda OM, Sammut SJ, Seoane JA, et al. Dynamics of breast-cancer relapse reveal late-recurring ER-positive genomic subgroups [published online March 13, 2019]. Nature. doi: 10.1038/s41586-019-1007-8