Preclinical Assessment of a MUC12-Targeted BiTE (Bispecific T-cell Engager) Molecule.

MUC12 is a transmembrane mucin that is highly expressed in >50% of primary and metastatic colorectal tumors. MUC12 is also expressed by normal epithelial cells of the colon and small intestine. Although MUC12 localization in normal epithelial cells is restricted to the apical membrane, expression in tumors is depolarized and shows broad membrane localization. The differential localization of MUC12 in tumor cells as compared with normal cells makes it a potential therapeutic target. Here, we evaluated targeting of MUC12 with a BiTE (bispecific T-cell engager) molecule. We generated a panel of proof-of-concept half-life extended (HLE) BiTE molecules that bind MUC12 on tumor cells and CD3 on T cells. We prioritized one molecule based on in vitro activity for further characterization in vivo In vitro, the MUC12 HLE BiTE molecule mediated T-cell-redirected lysis of MUC12-expressing cells with half-maximal lysis of 4.4 ± 0.9 to 117 ± 78 pmol/L. In an exploratory cynomolgus monkey toxicology study, the MUC12 HLE BiTE molecule administered at 200 µg/kg with a step dose to 1,000 µg/kg was tolerated with minimal clinical observations. However, higher doses were not tolerated, and there was evidence of damage in the gastrointestinal tract, suggesting dose levels projected to be required for antitumor activity may be associated with on-target toxicity. Together, these data demonstrate that the apically restricted expression of MUC12 in normal tissues is accessible to BiTE molecule target engagement and highlight the difficult challenge of identifying tumor-selective antigens for solid tumor T-cell engagers.

Pain sensation in human osteoarthritic knee joints is strongly enhanced by diabetes mellitus.

The major burden of knee joint osteoarthritis (OA) is pain. Since in elder patients diabetes mellitus is an important comorbidity of OA, we explored whether the presence of diabetes mellitus has a significant influence on pain intensity at the end stage of knee OA, and we aimed to identify factors possibly related to changes of pain intensity in diabetic patients. In 23 diabetic and 47 nondiabetic patients with OA undergoing total knee arthroplasty, we assessed the pain intensity before the operation using the "Knee Injury and Osteoarthritis Outcome Score". Furthermore, synovial tissue, synovial fluid (SF), cartilage, and blood were obtained. We determined the synovitis score, the concentrations of prostaglandin E2 and interleukin-6 (IL-6) in the SF and serum, and of C-reactive protein and HbA1c and other metabolic parameters in the serum. We performed multivariate regression analyses to study the association of pain with several parameters. Diabetic patients had on average a higher Knee Injury and Osteoarthritis Outcome Score pain score than nondiabetic patients (P < 0.001). Knee joints from diabetic patients exhibited on average higher synovitis scores (P = 0.024) and higher concentrations of IL-6 in the SF (P = 0.003) than knee joints from nondiabetic patients. Multivariate regression analysis showed that patients with higher synovitis scores had more intense pain independent of all investigated confounders, and that the positive association between pain intensities and IL-6 levels was dependent on diabetes mellitus and/or synovitis. These data suggest that diabetes mellitus significantly increases pain intensity of knee OA, and that in diabetic patients higher pain intensities were determined by stronger synovitis.