Abstract 2773: A DNA-damaging lupus autoantibody synergizes with PARP inhibitors against DNA repair-deficient tumor cells

The lupus anti-DNA autoantibody 3E10 is a compelling candidate for development as a targeted therapy for DNA repair-deficient malignancies. 3E10 has previously been shown to localize to tumors due to its attraction to DNA released by dying cancer cells, penetrate into cell nuclei, inhibit DNA repair, and kill cancer cells with defects in homology-directed repair (HDR) of DNA double-strand breaks. A more potent derivative of 3E10 with increased affinity for DNA has been developed (referred to here as 3E10EN), and identification of optimal combination therapies with 3E10EN is needed to facilitate planning for upcoming clinical trials. In the present study, we found that 3E10EN increases the activity of the DNA repair enzyme poly (ADP-ribose) polymerase (PARP) in HDR-deficient cells and hypothesized that combination treatment with 3E10EN and PARP inhibitors (PARPi) would yield synergistic effects on HDR-deficient cancer cell survival.

PARP content and activity in HDR-deficient and proficient cells prior to and following treatment with 3E10EN was evaluated. 3E10EN did not impact PARP protein content but yielded a significant increase in pADPr signal in HDR-deficient cells, which suggests a compensatory increase in PARP activity in response to DNA damage accumulation in HDR-deficient cells. Combinations of 3E10EN and the PARPi olaparib were tested on a panel of HDR-deficient cells, and a matched pair of BRCA2-deficient and proficient DLD1 cells. Olaparib inhibited the increase in pADPr caused by 3E10EN, and colony formation assays analyzed by the Chou-Talalay method confirmed that 3E10EN and olaparib synergized against HDR-deficient cancer cells. Conversely, HDR-proficient cells were resistant to 3E10EN and olaparib combination treatment.

The original 3E10 is a murine antibody isolated from a lupus mouse model, and in preparation for its further development as a new drug we have recently designed Deoxymab 1 (DX1), a humanized version of 3E10EN. DX1 exhibits improved activity relative to the 3E10EN prototype, and when tested on a panel of HDR-deficient and proficient cells, DX1 and olaparib exhibited synergistic effects similar to that observed with the 3E10EN prototype.

In conclusion, we have found that both the prototype 3E10EN and humanized DX1 synergize with PARPi against HDR-deficient tumor cells. These findings provide the rationale for further studies to determine the potential for this approach to be translated into a clinically relevant therapeutic strategy.

Citation Format: Zahra Rattray, Jaymin M. Patel, Philip W. Noble, Valentina Dubljevic, Deanne L. Greenwood, James A. Campbell, James E. Hansen. A DNA-damaging lupus autoantibody synergizes with PARP inhibitors against DNA repair-deficient tumor cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2773.

Characterisation of proton pump antibodies and stomach pathology in gastritis induced by neonatal immunisation without adjuvant

It has previously been reported that neonatal BALB.D2 mice injected with native proton pump antigens without adjuvant develop an irreversible gastritis (Claeys et al, 1997). The ease of inititating gastritis in the neonate stands in contrast with the difficulty in initiating gastritis in adult mice that require repeated immunisation in adjuvant that is reversible following cessation of immunisation (Scarff et al, 1997). In view of these contrasting observations, we set out to ascertain whether we could confirm the observations in neonatal mice as well as further characterise the pathology and the autoantibody response. We found that neonatal gastritis-susceptible BALB/c mice (n=12), immunised with either pig or mouse gastric membranes in the absence of adjuvant, develop gastritis without circulating antibody to parietal cells detected by immunofluorescence, a hallmark of murine and human gastritis (Toh et al, 1997). However, mice immunized with pig gastric membranes (n=6) had circulating antibodies reactive by immunofluorescence to recombinant alpha and/or beta subunit of gastric H+/K+-ATPase expressed by insect cells (Sfalpha and Sfbeta). Four mice from this cohort with antibodies to Sfbeta also had reactivity to gastric H+/K+-ATPase by ELISA, and 3 immunoblotted the beta but not the alpha subunit of the ATPase. In the cohort of mice immunised with mouse gastric membranes (n=6), four produced antibodies reactive by immunofluorescence to Sfalpha, two of which were also reactive to Sfbeta and one developed antibodies detected by ELISA to gastric H+/K+-ATPase. However, no members of this cohort had antibodies reactive by immunoblotting to either the beta or alpha subunit of the ATPase. In all cases gastritic stomachs were characterised by areas deficient in ribosome-rich zymogenic cells and marked reductions in H+/K+-ATPase-positive parietal cells. Metaplasia detected by Maxwell stain, as clusters of mucus-producing cells throughout gastric units, were non-reactive to stomach mucin autoantibody suggesting the mucins comprise other and/or aberrant form(s). Compared to our previous observations in adult mice, our present data confirms that gastric autoimmunity is more readily induced in the neonate than the adult. Our data also affirms that while the neonatal immune system can mount a damaging inflammatory cellular immune response to gastric antigens, it develops an altered antibody response.