145P - Selecting patients with metastatic colorectal cancer for treatment with temozolomide using proteomic analysis of MGMT

Abstract

Background: Temozolomide (TMZ) is a standard treatment for melanoma and glioblastoma and it has shown limited but encouraging activity in patients with metastatic colorectal cancer (mCRC). In multiple cancer types, the DNA repair protein O6-methylguanine-DNA methyltransferase (MGMT) is a resistance marker for TMZ; MGMT promoter methylation is associated with loss of MGMT expression and response to TMZ. We hypothesized that mCRC patients whose tumors expressed quantities of MGMT protein below a pre-defined cutoff would have better outcomes on TMZ than patients with MGMT expression above the cutoff. To test our hypothesis, we assessed MGMT by mass spectrometry in the tumor samples of patients with refractory mCRC and MGMT promoter methylation receiving TMZ. Methods: Archived formalin-fixed, paraffin-embedded tissue sections were obtained from 24 patients from two phase 2 trials. A pathologist marked the tumor areas, which were microdissected and solubilized. In each tumor sample, multiple protein biomarkers including MGMT were quantified with selected reaction monitoring mass spectrometry. An MGMT cutoff of 200 amol/ug was based on the limit of quantitation from a concentration curve. The Mantel-Cox log-rank and the Fisher’s exact tests were used for survival comparisons. Results: MGMT protein was detected in 13 of 24 (54.2%) colorectal tumor samples (range: 229.3-784.8 amol/µg). The overall response rate was 29%. Patients with MGMT protein levels below a cutoff of 200 amol/ug (n = 11) had a notably higher response rate than patients with MGMT levels above the cutoff (64% vs. 0%; p = 0.001 Fisher’s test). Also a longer progression-free survival was observed (4.3 vs. 1.6 months, HR = 0.36, 95% CI = 0.13-1.10, p = 0.054). Results for overall survival were consistent but not statistically significant (8.9 vs 6.9 months, HR = 0.55, p = 0.221). Conclusions: Patients with mCRC whose tumors expressed low or undetectable levels of MGMT protein had better outcomes following TMZ treatment than their counterparts. Quantitative proteomic analysis of MGMT could potentially be used to select CRC patients for TMZ treatment. The results of validation studies are forthcoming. Legal entity responsible for the study: NantOmics Funding: NantOmics Disclosure: S. Schwartz, F. Cecchi, Y. Tian, K. Scott, T. Hembrough: Employee at NantOmics. All other authors have declared no conflicts of interest.