Abstract P4-02-02: Magnetic resonance spectroscopy based total choline estimation: Does it have a role in the differentiation of grade II breast cancers into high and low risk groups?

Abstract

Purpose: Grade II breast cancers have been clinically attributed either to low risk or high risk groups based on the degree of cellular proliferation assessed by genomic grading index [1] or Ki67 labeling index. This differentiation is clinically important as the medical management of these two groups are different. Non-invasive MR Spectroscopy (MRS) based choline quantification enables assessment of cellular proliferation in-vivo. The aims of our study were to check whether (1) total choline (tCho) quantified with MRS correlated with tumor grading and (2) in grade II tumors, the differentiation into high and low risk groups could be made based on tCho values. Methods: The preliminary results concerning the first 14 patients included in a prospective ongoing study are discussed here. Patients who had tumor of at least 1.5cm dimension in all the three planes and who underwent mastectomy or wide excision were studied on a 1.5 T scanner (Philips Intera, Nederland) by in-vivo MR mammography protocol including MRS, which is a PRESS single voxel sequence (TE/ 270 ms, TR 2000ms). Post-processing was done using jMRUI 4.0. The AMARES algorithm was used to quantify tCho using water as the internal reference [2]. Operated tumor specimens were immunostained for Ki67 with MIB-1 antibody. Estimated tCho values were correlated with tumor grades and Ki67 index. Results: Among the 14 patients, 8 were grade II and 6 were grade III tumors. Table 1: Patient/tumor characteristicsAge (years)[Cho] (mmol/kg)Size (cm)GradeKi67 (%)ERPRHER2 Gene550,18811.7251++_671,13413.126,7+++580,00001.5210++_730,77154.1250__+720,00002.125++_610,00002.9250+__550,05054.625++_620,00003.0225++_540,07892.6326++_612,35772.8350++_620,45312.0330++_740,71802.1330___622,51252.8361++_710,60333.33NA__+(NA: not available) Ki67 could not be assessed in one grade III sample as it had extensive necrotic tissue. Strongly elevated tCho with a maximum of 2.53 mmol/kg was measured in grade III tumors. Mean tCho values were significantly different between Grade II and III tumors (t test, p = 0.05). There is a linear correlation between tCho and tumor grades (r = 0.54, p <0.05). There was a general correlation between Ki67 index and tCho, significant only in Grade III tumors (r = 0.98, p< 0.05). However, in grade II tumors, by taking the mean tCho (0.234mmol/kg) as the cut-off value, a rough delineation could be observed: a group of tumors (n = 6) with absent or very low(< 0.2mmol/kg) tCho and another group with a detectable tCho (n = 2). These two cases with detectable tCho were HER2 amplified. Conclusion: This feasibility study indicates significant correlation of tCho determined by MRS with tumor grades and a relatively weak correlation with Ki67 labeling index. This in turn offers a utilizing basis for further exploration of MRS based tCho estimation in the accurate in vivo assessment of breast tumor grades. These are preliminary results of an on-going study which will be further validated against the genomic grade index in a larger number of patients. References :[1]Ignatiadis M et al. Pathobiology.2008[2]Begley JKP et al. Breast Cancer Research 2012. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-02-02.