CUB Domain-Containing Protein 1 (CDCP1) is a rational target for the development of imaging tracers and antibody-drug conjugates for cancer detection and therapy

Rationale: An antibody-drug conjugate (ADC) is a targeted therapy consisting of a cytotoxic payload that is linked to an antibody which targets a protein enriched on malignant cells. Multiple ADCs are currently used clinically as anti-cancer agents significantly improving patient survival. Herein, we evaluated the rationale of targeting the cell surface oncoreceptor CUB domain-containing protein 1 (CDCP1) using ADCs and assessed the efficacy of CDCP1-directed ADCs against a range of malignant tumors. Methods: CDCP1 mRNA expression was evaluated using large transcriptomic datasets of normal/tumor samples for 23 types of cancer and 15 other normal organs, and CDCP1 protein expression was examined in 34 normal tissues, >300 samples from six types of cancer, and in 49 cancer cell lines. A recombinant human/mouse chimeric anti-CDCP1 antibody (ch10D7) was labelled with 89Zirconium or monomethyl auristatin E (MMAE) and tested in multiple pre-clinical cancer models including 36 cancer cell lines and three mouse xenograft models. Results: Analysis of CDCP1 expression indicates elevated CDCP1 expression in the majority of the cancers and restricted expression in normal human tissues. Antibody ch10D7 demonstrates a high affinity and specificity for CDCP1 inducing cell signalling via Src accompanied by rapid internalization of ch10D7/CDCP1 complexes in cancer cells. 89Zirconium-labelled ch10D7 accumulates in CDCP1 expressing cells enabling detection of pancreatic cancer xenografts in mice by PET imaging. Cytotoxicity of MMAE-labelled ch10D7 against kidney, colorectal, lung, ovarian, pancreatic and prostate cancer cells in vitro, correlates with the level of CDCP1 on the plasma membrane. ch10D7-MMAE displays robust anti-tumor effects against mouse xenograft models of pancreatic, colorectal and ovarian cancer. Conclusion: CDCP1 directed imaging agents will be useful for selecting cancer patients for personalized treatment with cytotoxin-loaded CDCP1 targeting agents including antibody-drug conjugates.

ch10D7. After seven days of culture, the conditioned media was centrifuged at 4000g for 10 min at 4°C then the supernatant was filtered (0.22 µm pore size). Antibody was purified from the clarified media by immunoaffinity chromatography on an ÄKTA pure protein purification system (Cytiva, Marlborough, MA, USA) using 5 a ml HiTrap MabSelect SuRe column (Cytiva), then desalted on a HiPrep 26/10 desalting column (Cytiva). After separation of the protein isolate by electrophoresis, ch10D7 purity was quantified by densitometric analysis of the Coomassie stained SDS-PAGE gel and its concentration was determined using a Nanodrop spectrophotometer (Thermofisher Scientific).

Antibody labelling with Atto-550
Labelling of human IgG1k, 10D7 and ch10D7 with fluorophore Atto-550 was performed following the instructions of the manufacturer. Briefly, antibodies were diluted to 2 mg/mL in PBS and incubated with the reactive Atto-550 dye (solubilized in DMSO) for 2 h at room temperature in the dark with gentle agitation. Labelled antibodies were separated from free excess dye using a gel filtration column with a size exclusion limit of 5 kDa. Columns were equilibrated with PBS before sample loading then elution of the labelled antibody with PBS. Eluted antibody concentrations were determined by BCA assay.

Immunofluorescent analysis
TKCC05 cells plated in 96 wells plate (5,000 cells per well) were washed with PBS then incubated in complete medium containing 5 µg/ml of IgG-550, 10D7-550 or ch10D7-550. Cells were fixed at defined time points (5 -120 min) using 4% PFA solution (15 min at room temperature), PBS washed and incubated with a solution of PBS containing DAPI (1:10,000) and 000) for 30 min at 4°C. After three washes, cells were imaged using an inverted fluorescent microscope.

Lysate preparation, immunoprecipitation and Western blot analysis
Cells were lysed in RIPA buffer containing EDTA-free Complete protease inhibitor (1x), sodium vanadate (2 mM) and sodium fluoride (10 mM anatomical pathologist (CES) blinded to clinical data and scored semi quantitatively as described previously [12,15]. For normal tissues, each core was assigned a score for staining intensity only (0, no staining; 1, weak; 2, moderate; or 3 strong). For cancer tissues, each core was assigned a score based on staining intensity (0, no staining; 1, weak; 2, moderate; or 3, strong) multiplied by the percentage of positive cells (in 20% increments) to produce a combined score that was assigned an integer value of 0 (combined score 0), 1 (combined score 1 to 100), 2 (combined score 101 to 200) or 3 (combined score 201 to 300).

In vitro ADC cytotoxicity assays
To assess the cytotoxic activity of IgG-MMAE, 10D7-MMAE and ch10D7-MMAE, cells (5,000/well) were plated in 96-well plates and allowed to attach for 24h [12,15]. Cells were then treated for 6 h with respective ADC at indicated concentration before to be washed with complete medium and incubation for another 72 h. Relative cell viability was then measured by adding CellTiter AQueous One Solution Reagent to each well and measuring absorbance at 490 nm.
For colony forming assays, cells were plated in 6 wells plate at a density of 100,000 cells per well and allowed to attach for 24h. Cells were then treated for 6 h with respective ADC at indicated concentration before to be washed with complete medium, detached using a non-enzymatic solution and centrifuge at 370g for 5 min.
Cell pellets were resuspended at a density of 500 cells/ml in complete medium and seeded in 24 wells plate. After 10 -14 days, cells were fixed using 4% PFA solution and stained with 4% crystal violet solution in methanol (w/v).

In vivo testing of ADCs
In vivo assessment of ADC efficacy was performed on luciferase labelled TKCC2.1 PDAC cells, HEY ovarian cancer cells, and HCT116 colon cancer cells, with bioluminescent tumor burden measured weekly, like what was outlined previously [12,15]. For the TKCC2.1 PDAC model, cells (1,000,000/mouse) were injected subcutaneously into the left flank of female NSG mice. Tumor volume was measured twice weekly (caliper measurement) and treatment started once tumors reached 200 mm 3 , at which point mice were randomised into groups (8-10 mice per group) with equivalent tumor volumes. Mice were treated with either vehicle (PBS, iv, 100µl, fortnightly), gemcitabine chemotherapy (i.p., 125 mg/kg weekly), or ADC (IgG-MMAE, 10D7-MMAE or hu/mu 10D7-MMAE, iv, 5mg/kg, fortnightly). Once mice in any group required euthanasia due to disease burden, treatment have been stopped and survival of mice have been followed.
For HEY ovarian cancer and HCT116 colon cancer experiments, female and male NSG mice respectively were injected intra-peritoneally with luciferase labelled HEY or HCT116 cells (100,000 cells per mouse). Tumors were allowed to grow for one week, after which time mice were randomised to groups (10 mice per group) with equivalent tumor burdens, and treatment begun with vehicle (PBS, iv, 100µl, fortnightly), chemotherapy (125 mg/kg weekly carboplatin for HEY cells I.P injection, 100mg/kg weekly 5FU for HCT116 cells I.P injection ), or ADC (IgG-MMAE, 10D7-MMAE or hu/mu 10D7-MMAE, iv, 5mg/kg, fortnightly). Once mice in any group required euthanasia due to disease burden, treatment have been stopped and survival of mice have been followed. Additionally, for HCT116 mice, when the first mouse was culled (Day 24), we euthanized 3 mice per group to assess metastatic burden in the mesenteric membrane, a common site of intraperitoneal colorectal cancer metastasis, using a previously established protocol [39]. Mesenteric tumor burden was quantified from examination of the right colonic mesentery and quantified using ImageJ software