In vitro and in vivo activity of the low-immunogenic antimesothelin immunotoxin RG7787 in pancreatic cancer

Abstract

Pancreatic ductal adenocarcinoma (PDAC) has a dismal prognosis, and new therapies are needed. RG7787 is a novel low-immunogenic antimesothelin recombinant immunotoxin (RIT), engineered to overcome the limitations of SS1P, a RIT now in clinical trials. In vitro activity was evaluated on five established PDAC cell lines (KLM-1, AsPC-1, BxPC-3, Panc 3.014, and PK-1) and on PDAC cells directly established from a patient tumor (GUMC108). RG7787 had subnanomolar IC50s in most cell lines, and was significantly more active than SS1P in GUMC108, KLM-1, and Panc 3.014 cells. GUMC108 was most sensitive, with RG7787 killing >99% of the cells. In a subcutaneous KLM-1 xenograft mouse model, two cycles of 3 × 2.5 mg/kg RG7787 QOD combined with two cycles of 1 × 50 mg/kg paclitaxel induced near-complete responses, with all tumors regressing below 5 mm(3) within 30 days after therapy was initiated (>95% decrease) and no significant growth increase for at least another 3 weeks. RG7787 alone gave limited but significant regressions and paclitaxel by itself arrested tumor growth. Quantifying the uptake of Alexa Fluor 647-labeled RG7787 in tumors showed that the RIT reached only 45% of KLM-1 cells, accounting in part for the limited responses. Paclitaxel did not improve RG7787 uptake, which thus cannot explain the beneficial effect of the combination therapy. In conclusion, RG7787 has high cytotoxic activity on PDAC cell lines as well as on primary patient cells. In vivo, this novel RIT gives durable near-complete tumor responses when combined with paclitaxel. RG7787 merits further evaluation for the treatment of PDAC.

Figure 1. Structural models of recombinant anti-mesothelin immunotoxins

Structural models of SS1P [SS1(dsFv)-PE38]and RG7787 [huSS1(Fab)-LR-GGS-LO10-PE24] are shown. Cartoons are created using VMD (34), based on the X-ray crystal structure of PE (Protein Data Bank code: 1IKQ). Models are hypothetical only and do not represent actual structural determinations. Native PE consists of three structural domains organized from a single polypeptide sequence. SS1P includes a 38-kDa PE truncation (PE38) that contains deletions in domain Ia and domain Ib. The model of SS1P shows the anti-mesothelin dsFv on the left and P38 on the right. RG7787 is a re-engineered version of SS1P. Modifications include removing the bulk of PE domain II, leaving a furin cleavage site, and adding a GGS peptide linker after the furin cleavage site. RG7787 is further optimized by replacing the mouse anti-mesothelin Fv (SS1) with a humanized anti-mesothelin Fab (huSS1) and by introducing seven mutations (R505A, R427A, R490A, R467A, D463A, R458A, and R538A) in the catalytic domain III of PE to silence B-cell epitopes.

Figure 2. Cytotoxicity of RG7787 versus SS1P in PDAC cell lines

A: Cell viability assays show that RG7787 is more cytotoxic than SS1P in the established cell line KLM-1 and primary GUMC108 cells. Cells were incubated for 72 hrs with SS1P orRG7787, and growth inhibition was evaluated with an ATP viability assay. Data is normalized between D-PBS HSA 0.2%- and staurosporine-treated controls. B: Viable cell counts done in triplicate of SS1P- and RG7787-treated KLM-1 and GUMC108 cells. Day 1 is just before addition of the RITs, and day 4 equals 72 hrs of treatment. C–D: Bright-field microscopy pictures of KLM-1 and GUMC108 confirm that RITs induce cell death. Cells were incubated with 100 ng/ml of SS1P or 117 ng/ml RG7787 for 72 hrs. Pictures (10X) were taken from identical locations within the wells at each time point. Representative series of pictures are shown. At day 4, floating dead cells were washed out prior to taking pictures. The cell count data (B) quantifies the visual observations.

Figure 3. Cellular uptake and protein synthesis inhibition of RG7787 versus SS1P in PDAC cell lines

A: In vitro cellular uptake of SS1P- and RG7787-Alexa647 is similar in KLM-1. Flow cytometry was used to obtain geomean fluorescence intensities (GFI) at each time point. B: RG7787 inhibits protein synthesis inhibition more than SS1P in KLM-1 and GUMC108. Cells were incubated with 100 ng/ml of SS1P or 117 ng/ml RG7787 for 16 hrs. Protein synthesis was evaluated by measuring [³H] leucine incorporation. Data is normalized between D-PBS HSA 0.2%- and cycloheximide-treated cells.

Figure 4. RG7787 efficacy and uptake in a KLM-1 xenograft model

Athymic nude mice were subcutaneously injected with 4 × 10⁶ KLM-1 cells, and treatment was initiated when tumors reached a volume of 120–130 mm³. Panels A and B present anti-tumor activity, and panels C and D the uptake of RG778 in the KLM-1 tumors quantified by a tumor dissociation flow cytometry method. A: Tumor-bearing mice were treated with D-PBS HSA 0.2% iv (control, n = 3) and one cycle of 3 × 0.4 mg/kg SS1P QOD iv (n=3). Black arrows present the three SS1P injections. SS1P had no effect on tumor growth. B: Mice were treated with D-PBS HSA 0.2% iv (control, n = 5, white diamonds), two cycles of 3 × 2.5 mg/kg RG7787 QOD iv (n=6, black diamonds), two cycles of 1 × 50 mg/kg paclitaxel ip (n=6, grey diamonds), or two cycles of 3 × 2.5 mg/kg RG7787 QOD iv and 1 × 50 mg/kg paclitaxel ip combined (n=6, grey-black diamonds), with each dose of paclitaxel administered 1 day before the first of three doses of RG7787. Two cylces of the paclitaxel and RG77787 combination induced durable near-complete regressions. One mouse in the combination group has an ongoing complete response 9 months after therapy. Follow-up in the combination group is plotted as long as groups were complete (day 88), i.e. until no mice had to be euthanized in order to comply with humane endpoints. Grey arrows are the time points of the two paclitaxel injections, black arrows present the six RG7787 injections. C: RG7787 uptake in vivo is limited to half of the tumor cells. KLM-1 tumors were harvested 5 min, 45 min, 3 hrs, 6 hrs, 9 hrs and 14 hrs after iv injection of 2.5 mg/kg RG778-Alexa647. The number of KLM-1 cells that internalize RG778-Alexa647 increases over time, with a maximal uptake of 45% at 6 hrs. D: Mice were injected with 2.5 mg/kg RG778-Alexa647 one day and three days after the injection of a single dose of 50 mg/kg paclitaxel ip, or without prior injection of paclitaxel. KLM-1 tumors were harvested 3 hrs after injection. Paclitaxel had no significant effect on RG7787 uptake. All experiments were performed at least in duplicate.

Figure 5. Immunohistochemical evaluation of KLM-1 tumors

KLM-1 tumors were stained with H&E and cleaved caspase 3, a marker of cell death. A: In an untreated tumor, KLM-1 cells were organized in clusters surrounded by connective tissue, and limited cleaved caspase 3 staining was observed. One day after 50 mg/kg paclitaxel ip (1d+), tumor organization was disrupted and significant cleaved caspase 3 staining was observed. B: In tumors treated with two cycles of 3 × 2.5 mg/kg RG7787 iv or two doses of 50 mg/kg paclitaxel ip alone, the effect of treatment was obvious. Organization was disrupted, and empty spaces were present within the tumor. In the tumor treated with two cycles of the combination, there was a vast presence of connective tissue with very few nests of tumor cells left. Representative pictures are shown (10X).