Intratumoral anti-HuD immunotoxin therapy for small cell lung cancer and neuroblastoma

To assure that mouse mAb 16A11 was specific for HuD-antigen in both human and mouse HuD-positive cancer cells, we first tested the reactivity of the antibody to cell extracts obtained from SCLC, neuroblastoma, and extracts those extracted from leukemia and mouse T lymphoma control cell lines by Western blot. Figure 1 shows that human SCLC (NCI-H69, DMS79) and mouse neuroblastoma neuro-2a cell extracts strongly express HuD (~39-40 kD) whereas control cell lines (BW5147 and K562) failed to express HuD antigen at the expected molecular weight. All cell lines displayed faint nonspecific binding at ~64 kD, but lymphoma control cell line BW5147 showed stronger signal compared to all other cell lines.

The killing effect of BW-2 was quantified in HuD-positive NCI-H69, Neuro-2a, and HuD-negative K562 tumor cell lines. Figure 2 shows that the BW-2 construct killed the targeted NCI-H69 SCLC (Figure 2A) and neuro-2a (Figure 2B) cells with great potency (ED₅₀ < 0.5 μg/ml). In contrast, the saporin-streptavidin complex alone or mAb alone exhibited limited non-specific killing only at much higher concentrations (P < 0.05). The addition of non-biotinylated mAb at a 20-fold excess only partially blocked cell killing by BW-2. Adding either streptavidin or biotin alone to NCI-H69 or neuro-2a cell cultures failed to block cell killing induced by BW-2. Minimal non-specific cytotoxicity was induced in HuD-negative control cells, but again only at much higher concentrations of BW-2 (2–5 μg/ml) (Figure 2A). Trypan blue staining for cell death detection as well as the Cell Titer 96 Proliferation Assay were utilized to confirm cell viability and all tests showed similar results (data not shown).

Nude mouse models of human SCLC have been used extensively by many investigators [8],[9] and tumor xenografts in the nude mouse usually develop within 2 weeks after subcutaneous inoculation of NCI-H69 cancer cells. In our experiments, the tumor mass expanded to 300-500 mm³ over 2–3 weeks after s.c. implantation of 1 × 10⁷ viable tumor cells into the flanks of nude mice. Tumor-grafted mice were capable of surviving for 6–8 weeks before they were overwhelmed by the tumor burden (Figure 3A). In contrast to common belief, we found that after s.c. tumor implantation, significant tumor metastasis developed in multiple organs (lung, liver, and bone marrow) after the local tumor reached 1000–1500 mm³ and the metastasized cancer cells were easily detected by immunoreaction with biotinylated anti-HuD mAb and Cy3 fluorescent detection (Figure 3B).

To determine whether intratumoral BW-2 affected tumor growth, the tumor xenografted nude mice were treated once with immunotoxin (BW-2 at 1 mg/kg) injected directly into the s.c. tumor. Control tumor xenografted mice received equivalent subcutaneous injections of pure anti-HuD mAb alone, saporin toxin alone, or sham treatment with saline. The tumor shrinking effect of immunotoxin treatment was clearly visible as control mice continued to demonstrate rapid tumor volume expansion whereas the primary tumors in the BW-2 treated animals remained static or actually regressed for ~3 weeks (Figure 4A/B). The residual tumor eventually re-grew into a larger mass 6–9 weeks after one time local immunotoxin treatment.

We hypothesized that intratumoral immunotoxin therapy for SCLC or NB might be more effective in immunocompetent mice because the immunotoxin-induced tumor cell death would release abundant tumor antigens, which could potentially activate infiltrating immune cells locally to kill residual tumor cells and prevent distant metastasis. Because a mouse model of SCLC in immunocompetent mice is not available, the mouse model of HuD-positive neuro-2a NB in immunocompetent A/J mice was used for this study. The neuro-2a cell is a subclone of a murine neuroblastoma that was harvested from a spontaneously developed tumor in A/J mice. The neuro-2a implanted A/J mouse model has been used previously for studying efficacy of DNA vaccines [10]. As shown in Figure 5, more marked efficacy of BW-2 was observed in neuro-2a tumor allografted immunocompetent A/J mice. In BW-2 treated mice, a temporary loss of fur was observed at the injection site. Fur began to grow back within a few weeks after the final treatment. Otherwise, no other adverse effects were detected. Long-term follow-up over 7 weeks showed no local tumor relapse or distant metastasis in 80% of the BW-2 treated immunocompetent A/J mice. In contrast, the sham treated control mice had overwhelming local tumor growth. Additionally, multiple organs (brain, lung, liver, adrenal gland and bone marrow) had clearly detectable tumor metastases when examined either by direct tissue examination (Figure 5) or by immunohistochemistry.

Human SCLC (NCI-H69, DMS-79), human erythroleukemia (K-562), mouse NB (neuro-2a) and mouse T lymphoma (BW5147) cell lines were purchased from the American Type Culture Collection (ATCC; Rockville, MD). Cells were maintained in growth medium supplemented with 10% fetal calf serum (FCS) as recommended by the vendor. Saporin (SAP) and streptavidin-SAP were purchased from Advanced Targeting Systems (Santiago, CA) and stored at −80°C until use. Biotinylated and non-biotinylated mouse anti-human HuD mAb 16A11 (IgG2b) were purchased from Molecular Probes (Eugene, OR) [8]. Immunotoxin BW-2 was prepared by premixing biotinylated-16A11 mAb with streptavidin-SAP in equal molar concentrations.

Neuro-2a (mouse neuroblastoma) cells were lysed and total protein content was determined by measuring its absorbance. Aliquots of 50 μg total protein from each sample were subjected to SDS-polyacrylamide gel electrophoresis on a NuPAGE 4-12% Bis-Tris gel (Invitrogen, Carlsbad, CA) followed by transfer to a nitrocellulose membrane. The membrane was probed with primary antibodies anti-HuC/D (6ug/mL, Molecular Probes, Eugene, OR) followed by incubation with appropriate HRP-conjugated secondary antibodies (GE healthcare, Piscataway, NJ). The signal was visualized through the use of the ECL Plus Western Blotting Detection System (GE Healthcare) and exposure to ECL Hyperfilm (GE Healthcare). The membrane was stripped with Re-Blot Plus Strong (Chemicon, Temecula, CA).

Five ×10⁴ tumor cells (NCI-H69, DMS-79, neuro-2a or control K562 cells) in 100 ul of culture medium were placed in each well of 96-well plates. Log dilutions (0.005, 0.05, 0.5 and 5 μg/ml) of BW-2 were added to the cell culture wells. Equivalent molar concentrations of SAP, antibody alone as well as streptavidin alone served as controls. The cells were incubated at 37°C, 5%CO₂ for 72 h. Cell viability was assayed using a modified tetrazolium salt 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (Cell Titer 96 Cell Proliferation Assay kit, Promega Corp., Madison, WI) according to the manufacturer’s instructions. Tunnel or trypan blue staining was used on selected cell lines to confirm the findings. Each assay was performed in triplicate.

Male athymic nude mice at age 5–6 weeks (NCRNUM) were purchased from Taconic (Hudson, NY). Female A/J mice at 6–8 weeks were purchased from Jackson laboratory (Bar Harbor, ME). The mice were housed and maintained in specific pathogen-free conditions and the studies were conducted in accordance with the Animal Component of Research Protocol guidelines at the VA Medical Center, East Orange, NJ.

The mouse model of human SCLC was established by subcutaneous (s.c.) implantation of viable tumor cells (1 × 10⁷/100 μl PBS) into the flank of nude mice. Animals were monitored 5 days per week for tumor onset and growth. Tumor size was measured using calipers and final tumor volumes were calculated using a standard formula: length × width × height × 1/6 × π [9].

A mouse model of neuroblastoma was similarly established by s.c. implantation of mouse neuro-2a cells (1 × 10⁷/100 μl PBS) into the flank of immunocompetent female A/J mice at 6–8 weeks of age [10]. Tumor size was measured 5 days per week as above. When primary tumors reached approximately 150-200 mm³, mice (n = 5) received two intratumoral injections at weekly intervals of either BW-2 at a dose of 1 mg/kg or equivalent doses of PBS. Mice were monitored following treatments 5 times per week for tumor growth and general condition. Animals were sacrificed and perfused with 4% paraformaldahyde when tumor burden became too great.

When subcutaneous tumor nodules reached approximately 150–200 mm³ in size, the SCLC tumor xenografted nude mice (n = 6) received treatment with one time intratumoral immunotoxin at doses of 1 mg/kg, or equivalent doses of pure 16A11 mAb alone, saporin toxin alone, or sham treatment with saline. Mice were monitored 5 times per week by measuring tumor size, assessing general condition, weight changes, and presence of neurological symptoms post treatment.

Minor modifications of the treatment protocol were made for the NB allografted A/J mice (n = 5). When subcutaneous NB tumor nodules reached a size of 150–200 mm³, mice received two intratumoral immunotoxin treatments at weekly intervals. Mice receiving equivalent doses of mAb (16A11) alone, saporin toxin alone, or sham treatment with saline served as controls. Mice were then monitored for tumor growth and general condition as described above.

Local tumors and organs (lung, brain, liver, adrenal gland and bone marrow) were collected for histopathologic studies at serial time points post treatment. Standard hematoxylin/eosin (H&E) and immunohistochemical reactions on the tissue sections were performed with anti-HuD antibody (16A11) to determine the extent of tumor involvement. Tissue sections of GFP-labeled neuro-2a tumors or organs that exhibited metastasis were visualized directly under fluorescent microscopy. To quantify the immune cell infiltration within the local s.c. tumor sites, immunochemical reactions on the frozen tumor tissue sections were conducted with antisera (eBioscience, San Diego, CA) specific for CD4, CD8, CD11c and Foxp3+ following manufacturer’s recommendations.

Statistical analysis was performed using Instat software 3.0 (Graphpad Software, Philadelphia, PA). The composite data was analyzed by the Kruskal-Wallis one-way analysis of variance and the Mann–Whitney U test (two-tailed) was used to determine the significance of intergroup clinical response differences after treatment. A value of P < 0.05 was considered significant.