Novel Dual-Action Targeted Nanomedicine in Mice With Metastatic Thyroid Cancer and Pancreatic Neuroendocrine Tumors

Abstract

Background: The advantages of nanomedicines include preferential delivery of the payload directly to tumor tissues. CYT-21625 is the novel, first-in-class gold nanomedicine designed to target tumor vasculature and cancer cells by specifically delivering recombinant human tumor necrosis factor alpha (rhTNF) and a paclitaxel prodrug.

Methods: We analyzed TNF receptor expression in publicly available gene expression profiling data and in thyroid tissue samples. Mice with metastatic FTC-133 and 8505C xenografts and the MEN1 conditional knock-out mice were treated weekly with CYT-21625 and gold nanoparticles with rhTNF only (CYT-6091); controls included mice treated with either paclitaxel or saline. In vivo luciferase activity was used to assess the effects on tumor growth. Computed tomography, magnetic resonance imaging, and 18F-Fludeoxyglucose positron emission tomography were used to study tumor selectivity in mice with insulin-secreting pancreatic neuroendocrine tumors (PNETs). All statistical tests were two-sided.

Results: Anaplastic thyroid cancer (ATC) expressed statistically significantly higher levels of TNF receptor superfamily 1A and 1B messenger RNA (n = 11) and protein (n = 6) than control samples (n = 45 and 13, respectively). Mice (n = 5-7 per group) with metastatic ATC (P < .009) and FTC-133 xenografts (P = .03 at week 3, but not statistically significant in week 4 owing to reduced sample size from death in non-CYT-21625 groups) treated with CYT-21625 had a statistically significantly lower tumor burden. Treatment with CYT-21625 resulted in loss of CD34 expression in intratumoral vasculature, decreased proliferating cell nuclear antigen, and increased cleaved caspase-3. Intratumoral vascular leakage occurred only in mice with PNET and ATC treated with CYT-6091 and CYT-21625. CYT-6091 and CYT-21625 preferentially deposited in PNETs and statistically significantly decreased serum insulin levels (n = 3 per group, P < .001). There were no toxicities observed in mice treated with CYT-21625.

Conclusions: CYT-21625 is effective in mice with PNETs and metastatic human thyroid cancer with no toxicities. Thus, CYT-21625 should be studied in patients with advanced PNETs and thyroid cancer.

Figure 1.

Tumor necrosis factor (TNF) receptor expression in normal, benign thyroid tissue, and thyroid cancers. A)TNFRSF1A and TNFRSF1B messenger RNA expression in anaplastic thyroid cancer (ATC) (n = 11) and normal thyroid tissue (n = 49). TNFRSF1A protein expression in (B) normal thyroid and follicular adenoma, (C) papillary thyroid cancer and follicular thyroid cancer, (D) ATC, (F) 8505C and FTC-133 lung xenografts, and (G) pancreatic neuroendocrine tumors and normal pancreas. Immunohistochemistry score of TNFRSF1A expression in thyroid samples is shown in (E). Arrows indicate TNFRSF1A expression in intratumoral blood vessels. Arrow heads (D) indicate TNFRSF1A expression in cytoplasm. The boxes in (A) represent the interquartile range contaning the middle 50% of the data. The lines across the box plot represent the median. The error bars from the upper and lower ends of the boxes represent 1.5 times the interquatile range. Error bars in (E) represent standard deviation. The comparison of messenger RNA expression between groups was performed by the moderated F-statistic, which combines the t-statistics for all the pair-wise comparisons into an overall test of statistical significance for that gene. Two-side P values were adjusted for false discovery rate using the Benjamini and Hochberg method (27). Two-sided analysis of variance with post hoc tests was used to compare the immunohistochemistry scores. Scale bars = 50 µm. ATC = anaplastic thyroid cancer; FTC = follicular thyrid cancer; PNET = pancreatic neuroendocrine tumor; PTC = papillary thyroid cancer; TNFRSF1A and TNFRSF1B = tumor necrosis factor receptor superfamily, member 1A and 1B.

Figure 2.

In vivo whole-body luciferase activity in mice. A) Metastatic human anaplastic thyroid cancer (8505C) xenografts. P values indicate statistically significant differences when the luciferase activity in mice treated with vehicle (n = 7, black line) and CYT-6091 (n = 6, red line) was compared with CYT-21625 (n = 5, green line). B) Metastatic human follicular thyroid cancer (FTC-133) xenografts. P values indicate statistically significant differences when the luciferase activity in mice (n = 5 per group) treated with vehicle (black line), paclitaxel (blue line), and CYT-6091 (red line) was compared with CYT-21625 (green line). The two-sided Student t test was used to assess the difference in luciferase activity between groups.

Figure 3.

Hematoxylin and eosin (H&E), immunohistochemistry, and immunofluorescence staining of metastatic anaplastic thyroid cancer (8505C) xenograft and normal organs (heart, liver, and kidney) in mice treated with CYT-21625, CYT-6091, paclitaxel, and vehicle. A) CD34 expression in intratumoral blood vessels (arrows), (B) H&E staining of normal organs (arrows indicate gold nanoparticles deposit), (C) CD34 expression in vascular endothelium of normal organs (arrows), (D) the extravasation of fibrinogen in 8505C xenografts (arrows) and in (E) normal organs of mice treated with CYT-6091 and CYT-21625 (Merge (DAPI+Anti-fibrinogen)), (F) proliferating cell nuclear antigen in 8505C and FTC-133 xenografts, and (G) cleaved caspase-3 expression in 8505C and FTC-133 xenografts. The boxes in (A) represent the interquartile range contaning the middle 50% of the data. The lines across the box plot represent the median. The error bars from the upper and lower ends of the boxes represent 1.5 times the interquatile range. Two-sided analysis of variance with post hoc tests was used to compare the immunohistochemistry scores. Scale bars = 50 µm.

Figure 4.

In vivo evaluation of biodistribution of CYT-21625 by computed tomography (CT) and 18F-FDG positron emission tomography (PET) scan. CT shows increased gold particle density in the pancreatic region (circle) in the multiple endocrine neoplasia type 1 gene knock-out mice treated with CYT-6091 and CYT-21625 at 24 hours postinjection (top), but not in the mice treated with paclitaxel and vehicle control (bottom). 18F-FDG PET images demonstrate pancreatic neuroendocrine tumor location in the midabdominal area, corresponding with the area where gold particles deposited. CT = computed tomography; PET = positron emission tomography.

Figure 5.

Imaging vascular effects of CYT-21625 by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). The signal enhancement curve was used for quantifying the tumor vasculature (A, C, E, G) and control muscle vasculature (B, D, F, H) by DCE-MRI. The first six time points are pre–injection of gadolinium with diethylenetriaminepentacetate (Gd-DTPA); the next 80 are post–injection of Gd-DTPA. E(t) = [S(Ct)-S(0)]/S(0), where S(0) is the signal absence of Gd-DTPA (Ct = 0) and S(Ct) is the signal after injection of Gd-DTPA. The data are from the pancreatic and control muscle regions of multiple endocrine neoplasia type 1 gene knock-out mice (n = 3) at 24 hours postinjection (red curve) and baseline (green curve). PNET = pancreatic neuroendocrine tumor; RSE = Relative signal enhancement.

Figure 6.

Antitumor activity of CYT-21625 was evaluated in MEN1 KO mice. Serum insulin levels were measured using an enzyme-linked immunosorbent assay at baseline and 7 and 14 days after treatment with CYT-6091, CYT-21625, paclitaxel, or vehicle (n = 3 per group). One mouse each in the paclitaxel-treated and vehicle control (phosphate-buffered saline–treated) groups died. The statistical significance within each of the four groups was calculated compared with their respective baselines using a two-sided Student t test. Error bars represent standard deviation. At day 14, one mouse died in the paclitaxel group and one died in the vehicle control group.