Abstract P1-09-03: Relative radioresistency in triple negative tumors in the SweBCG91-RT randomized clinical trial

Abstract

Abstract Introduction: Breast-conserving surgery (BCS) with adjuvant whole breast radiation therapy (WBRT) is the standard treatment for a majority of early breast cancer patients. No predictive biomarkers for RT are in use and most patients are cured by surgery alone, and are thus over-treated. Further, some patients suffer a relapse despite WBRT, and may have benefited from mastectomy or more aggressive postoperative treatment. Gene expression tests can be used to predict risk of distant recurrence and effect of adjuvant systemic therapy, and can reveal the intrinsic subtype of the tumor. A surrogate method of determining intrinsic subtype based on high quality centralized immunohistochemistry (IHC) has been proposed with criteria set up by the St Gallen consensus group 2013. The intrinsic subtypes provide prognostic information and are treatment predictive for chemotherapy, but the predictive potential for WBRT has not been conclusively determined. Aim: To evaluate the effect of WBRT on ipsilateral breast tumor recurrence (IBTR), in patients with tumors of different intrinsic subtypes. Methods: Tumor tissue from FFPE blocks were collected from 1003 breast cancer patients with node negative, stage I-II disease, randomized to BCS with or without WBRT, in the randomized SweBCG RT-91 trial between 1991-1997. Systemic adjuvant treatment was administered according to regional guidelines, but was sparsely used. Median follow-up was 15.2 years. Tissue microarrays were constructed and stained for estrogen receptor (ER), progesterone receptor (PgR), human epidermal growth factor receptor 2 (Her2) and Ki-67. SISH was used to determine amplification of samples scored 2+ for Her2. Centralized evaluation was performed by two pathologists subspecialized in breast pathology. Endpoint IBTR within 10 years was considered with a cumulative incidence and competing risks approach. P-values were calculated with the cause-specific logrank test and hazard ratios (HR) with cause specific Cox regression. Multivariate models, with or without an interaction term between subtype and WBRT, were compared to formally test if the effect of RT differs between subtypes. Results: We were able to stain and score 958 out of 1003 tumors. These were classified as Luminal A-like (n=554), Luminal B-like (Her2-negative, n=259), triple negative (n=81) and Her2-positive (any ER status, n=64). WBRT reduced the frequency of IBTR for Luminal A-like tumors (19% vs 9%, HR 0.46 (0.28-0.74), p=0.001), Luminal B-like tumors (24% vs 8%, HR 0.30 (0.14-0.61), p<0.001) and triple negative tumors (21% vs 6%, HR 0.25 (0.05-1.12), p=0.05), but not for Her2-positive tumors (15% vs 19%, HR 1.29 (0.38-4.4), p=0.69). However, the overall difference in WBRT effect between subtypes was not formally statistically validated (p=0.17). Conclusions: We found that WBRT reduced IBTRs among the Luminal A, Luminal B, and the triple negative subgroups, but not in the Her2-positive subgroup. Thus, intrinsic subtyping by IHC may give information on how tumors respond to adjuvant WBRT. Additional studies are required and it remains to study the effect on breast cancer specific survival. Citation Format: Sjöström M, Lundstedt D, Hartman L, Holmberg E, Kovács A, Malmström P, Niméus E, Werner Rönnerman E, Fernö M, Karlsson P. Relative radioresistency in triple negative tumors in the SweBCG91-RT randomized clinical trial [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-09-03.