70-Gene Signature and Tumor-Stroma Ratio Select Different Groups of Patients with Breast Cancer at Risk for Recurrence

Author(s): Marloes A Smit, Julia EC van Steenhoven, Rob AEM Tollenaar, Thijs van Dalen, Paul J van Diest, Wilma E Mesker

Background: Strategies for breast cancer care are improving. However, there is a clinical need for a better selection of patients who are at risk for recurrence and benefit from treatment with chemotherapy. The 70-gene signature (70-GS) guides chemotherapy decision making based on the mRNA microarray assessment but is expensive. The tumor-stroma ratio (TSR) is a morphological and cheap biomarker based on the amount of stroma in the primary tumor. In the present study the association between the 70-GS and TSR was investigated.

Materials and methods: Frozen tumor tissue sections of 102 ER+/Her2- patients were send to Agendia for 70-GS analysis in an earlier study. The results of the 70-GS were available, and hematoxylin and eosin stained sections of the 102 tumors were scored for TSR. Concordance between the TSR and the 70-GS was analyzed. No data of patient outcome is available.

Results: Of the patients in the 70-GS high risk category (n=46), 23 patients (50%) were classified into the stroma-high group according to previously established thresholds. 55% (n=31) of the low-risk 70-GS patients, had a stroma-low profile. The overall concordance between the 70-GS and TSR was 53% and no association was observed (Chi-square: p=0.917). Patients with a 70-GS high-risk test result had a significantly higher tumor grade and stroma-high patients were significantly older and had larger tumors.

Conclusion: In conclusion, no association between the 70-GS and TSR was found. This suggests that both methods select different groups of patients at risk for recurrence, and may have additional prognostic value.