Therapeutic effect of cisplatin given with a lymphatic drug delivery system on false-negative metastatic lymph nodes

Abstract

Systemic administration of drugs into the blood circulation is standard treatment for prevention of metastasis. However, systemic delivery cannot maintain sufficiently high concentrations of anticancer drugs in lymph nodes (LN). Here, we show that giving cisplatin (CDDP) using a lymphatic drug delivery system (LDDS) has the potential to treat false-negative metastatic LN. We found that in MXH10/Mo-lpr/lpr mice, which develop systemic swelling of LN up to 10 mm in diameter, accumulation of indocyanine green (ICG), which has a similar molecular weight to CDDP, in a target LN was greater for lymphatic delivery of ICG than for systemic (i.v.) administration. Furthermore, CDDP administration with a LDDS inhibited tumor growth in false-negative metastatic LN and produced fewer adverse effects than systemically given CDDP. We anticipate that drug delivery using a LDDS will, in time, replace systemic chemotherapy for the treatment of false-negative metastatic LN.

Keywords: Cisplatin; MXH10/Mo-lpr/lpr mouse; lymph node metastasis; lymphatic drug delivery system; lymphatic network.

Figure 1

Biodistribution of indocyanine green (ICG) after systemic administration or lymphatic injection into the accessory axillary lymph node (AALN). (A) In vivo whole body fluorescence images to evaluate the distribution of ICG in mice at different time points post‐injection. The lymphatic drug delivery system (LDDS) group shows high fluorescence in the axillary area (white arrows) for 24 h after ICG injection. In contrast, fluorescence was not detected in the axillary area in the i.v. injection (IV) group, and ICG seemed to accumulate in the liver. (B) Ex vivo fluorescence images at 24 h post‐injection of ICG. (a) Ex vivo fluorescence images of each organ. For the LDDS group (n = 7), ICG fluorescence was detected in the proper axillary lymph node (PALN) and AALN. However, for the IV group (n = 4), fluorescence levels in the PALN and AALN were below the limits of detection. (b) Ex vivo fluorescence intensity of each organ normalized to its weight. Statistically significant differences between the LDDS and IV groups were found for the liver, PALN and AALN (Mann–Whitney U‐test: *P < 0.05, liver; *P < 0.05, PALN; **P < 0.01, AALN). (c) In vitro fluorescence intensity of plasma. There was no significant difference between the IV (n = 4) and LDDS (n = 4) groups (Mann–Whitney U‐test).

Figure 2

Antitumor effect of cisplatin (CDDP) in the metastatic proper axillary lymph node (PALN). (A) Induction of metastasis in the PALN. Tumor cells were injected into the subiliac lymph node (SiLN) to induce metastasis to the PALN by lymphatic vessels. (a) In vivo bioluminescence imaging. Luciferase activity in the PALN was measured at days 7 and 14 and every 3 days after day 14. Metastasis was considered to have been successfully induced when PALN luciferase activity exceeded the background level of controls (1 × 10⁶ photons/s). (b) Luciferase activity increased with time in both the SiLN (n = 3, 3, 3, 7, 7, 5, 2 for days 7, 14, 17, 20, 23, 26, 29, respectively) and PALN (n = 3, 7, 3, 7, 7, 3, 3 for days 7, 14, 17, 20, 23, 26, 29, respectively), but was higher in the SiLN. (B) In vivo bioluminescence imaging after treatment intervention. (a) In vivo bioluminescence imaging. The day on which PALN luciferase activity reached 1 × 10⁶ photons/s was defined as day –1^T, and treatment was given on the following day (day 0^T). Mice were divided into lymphatic drug delivery system (LDDS) and i.v. injection (IV) groups, and the LDDS group was subdivided into three groups: 0 μg/g CDDP (saline alone, n = 5), 0.5 μg/g CDDP (n = 5) and 5 μg/g CDDP (n = 5). For the IV group, 5 μg/g CDDP was injected into the tail vein (n = 5). Among the various treatment groups, the LDDS 5 μg/g CDDP group showed the highest inhibition of tumor growth in the PALN. (b) Changes in luciferase activity in PALN. In vivo luciferase activity after treatment, normalized in each group to the value at day –1^T. Each group, n = 5. There was no statistical significance between each group on days 3^T, 6^T and 9^T (Kruskal‐Wallis test). (C) Ex vivo bioluminescence imaging at day 9^T after treatment. (a) Each organ was harvested at day 9^T after treatment and luciferase activity measured. There was no detectable luciferase activity in PALN in the LDDS 5 μg/g CDDP group. There was no detectable luciferase activity in accessory axillary lymph node (AALN) or lungs in all groups. (b) Ex vivo luciferase activity at day 9^T after treatment. Each group, n = 5. Kruskal‐Wallis test, PALN: *P < 0.05, LDDS 0 μg/g CDDP vs LDDS 5 μg/g CDDP.

Figure 3

Evaluation of proper axillary lymph node (PALN) and accessory axillary lymph node (AALN) volumes after treatment. (A) Changes in the volume of PALN measured using a high‐frequency ultrasound (HFUS) imaging system. (a) Images of PALN at day 9^T after treatment. (b) Comparison of PALN volume between the inoculation day and day 0^T. In all groups, there were no significant increases in volume of PALN, indicating that metastatic PALN met the criteria for a false‐negative metastatic lymph node (LN) (Wilcoxon signed rank test). (c) Effect of treatment on PALN volume. There were no significant differences in PALN volume between groups on the inoculation day and at day 0^T. In the control group, PALN volume increased about two‐fold from the inoculation day to day 9^T. The lymphatic drug delivery system (LDDS) 5 μg/g cisplatin (CDDP) group exhibited an inhibition of PALN volume growth following treatment, with a statistically significant difference observed between the LDDS 0 μg/g CDDP and LDDS 5 μg/g CDDP groups (Kruskal‐Wallis test: *P < 0.05, 0 μg/g CDDP vs 5 μg/g CDDP). (B) Changes in volume of AALN measured using a HFUS imaging system. (a) Images of AALN at day 9^T after treatment. (b) Effect of treatment on AALN volume. There were no significant changes in AALN volume in the LDDS 0 μg/g CDDP and LDDS 0.5 μg/g CDDP groups. However, the LDDS 5 μg/g CDDP and i.v. injection (IV) (5 μg/g CDDP) groups showed a decrease in AALN volume after treatment. Statistically significant differences were observed between the LDDS 0 μg/g CDDP group and the LDDS 5 μg/g groups (Kruskal‐Wallis test: *P < 0.05, 0 μg/g CDDP vs 5 μg/g CDDP).

Figure 4

Histological analysis. (A1–A4) lymphatic drug delivery system (LDDS) 0 μg/g cisplatin (CDDP) group. (B1–B4) LDDS 0.5 μg/g CDDP group. (C1–C4) LDDS 5 μg/g CDDP group. (D1–D4) i.v. injection (IV) (5 μg/g CDDP) group at day 9^T. (A1–A3, B1–B3, C1–C3, D1–D3) proper axillary lymph node (PALN) stained with H&E, anti‐CD31 antibody or anti‐LYVE‐1 antibody. In the LDDS 0 μg/g CDDP (A3), LDDS 0.5 μg/g CDDP (B3) and IV (D3) groups, tumor had spread widely throughout the PALN despite treatment. In the LDDS 5 μg/g CDDP group (C3), tumor cells were observed only in lymphatic vessels. In all groups, lymphatic vessels had been expanded by tumor growth (A3, B3, C3, D3). LEC, lymphatic endothelial cells; T, tumor cells. Scale bar, 50 μm. (A4, B4, C4, D4) H&E staining of the kidney. Atrophy in the proximal convoluted tubules was observed in kidney sections from the IV group. Scale bar, 20 μm.