EZH2 Expression and Response to Neoadjuvant Endocrine Therapy in Estrogen Receptor Positive Invasive Breast Cancer

ABSTRACT

Background: Neoadjuvant endocrine therapy (NET) is used to treat estrogen receptor (ER) positive invasive breast cancer (IBC). Tumors with Ki67 >10% after 2-4 weeks of NET are considered resistant to endocrine therapy (ET). Although high baseline Ki67 >30% is associated with higher level of ET resistance, many of these patients will still respond to NET. During the COVID- 19 pandemic NET is increasingly being used to defer surgery, hence better and more easily accessible biomarkers are needed to predict likelihood of response to NET. EZH2 is a oncoprotein which can be easily evaluated by immunohistochemistry and overexpression of EZH2 in ER+ IBC has been linked to resistance to ET. We examined the potential utility of EZH2 to predict response to NET . Design: We identified 34 pts with ER+ IBC of ductal or lobular type who received NET. Ki67 IHC was evaluated on pretherapy biopsies and post-therapy resections and scored according to guidelines of the International Ki67 Working Group with a global weighted score. We quantified EZH2 nuclear expression in pretherapy biopsies using a score which multiplied intensity (0=negative,1=weak,2=moderate,3=strong) by % of cells staining at each intensity X100 . Ki67 post therapy <=10% was considered endocrine responsive. Pretherapy Ki67 was dichotomized into >30% and <=30%. Results: The pt age range was 48 to 85 yrs (mean 64 yrs and median 65 yrs). All IBCs had ER expression levels >=80% and aromatase inhibitor was the most frequent NET (76%). Duration of NET ranged from 2-24 months (median 6 months and mean 6.5 months). Twenty pts had a pretherapy Ki67 < or =30 % and 14 had a pretherapy Ki67 >30%. Amongst the 20 pts with pretherapy Ki67<=30%, 11 were endocrine responsive and had a mean pretherapy EZH2 of 88 and median of 85 ( range 2-150) whereas 9 were resistant with a median pretherapy EZH2 of 108 and a median of 104 ( range 42-130). There was no significant difference in mean EZH2 score between groups ( t test p=.0.3971). Amongst the 14 pts with pretherapy Ki67>30% 5 were endocrine responsive and had a mean pretherapy EZH2 of 91 and median EZH2 of 93 (range 45-130) whereas 9 were resistant and had a mean pretherapy EZH2 of 186 and median of 178 ( range 120-240) There was a significant difference in mean EZH2 score between groups ( t test p=.0006). There was a significant association between pretherapy EZH2 score >130 and resistance to NET in pts with Ki67>30% pretherapy (p=.0030). Conclusions: In our pilot study EZH2 protein expression levels were significantly associated with response to NET in pts with high risk (Ki67>30%) ER+ IBC ;high EZH2 expression (>130) in IDC in pretherapy core biopsies was associated with resistance to NET in these pts. During the Covid-19 Pandemic , or in other situations where surgery might be deferred, our results suggest that EZH2 might be useful to predict tumor response to NET in high risk (Ki67>30%) ER+ IBC. (Table Presented).