Noninvasive detection of breast cancer lymph node metastasis using carbonic anhydrases IX and XII targeted imaging probes

Abstract

Purpose: To develop targeted molecular imaging probes for the noninvasive detection of breast cancer lymph node metastasis.

Experimental design: Six cell surface or secreted markers were identified by expression profiling and from the literature as being highly expressed in breast cancer lymph node metastases. Two of these markers were cell surface carbonic anhydrase isozymes (CAIX and/or CAXII) and were validated for protein expression by immunohistochemistry of patient tissue samples on a breast cancer tissue microarray containing 47 normal breast tissue samples, 42 ductal carcinoma in situ, 43 invasive ductal carcinomas without metastasis, 46 invasive ductal carcinomas with metastasis, and 49 lymph node macrometastases of breast carcinoma. Targeted probes were developed by conjugation of CAIX- and CAXII-specific monoclonal antibodies to a near-infrared fluorescent dye.

Results: Together, these two markers were expressed in 100% of the lymph node metastases surveyed. Selectivity of the imaging probes were confirmed by intravenous injection into nude mice-bearing mammary fat pad tumors of marker-expressing cells and nonexpressing cells or by preinjection of unlabeled antibody. Imaging of lymph node metastases showed that peritumorally injected probes detected nodes harboring metastatic tumor cells. As few as 1,000 cells were detected, as determined by implanting, under ultrasound guidance, a range in number of CAIX- and CAXII-expressing cells into the axillary lymph nodes.

Conclusion: These imaging probes have potential for noninvasive staging of breast cancer in the clinic and elimination of unneeded surgery, which is costly and associated with morbidities.

Figure 1

Expression of CA9 and CA12 mRNA in patient and unaffected tissue samples. (A) DNA microarray expression profile of CA9 and CA12 in breast tumor, lymph node positive and corresponding normal samples. Data are represented as mean ± s.d. Note the Log 10 scale. Asterisks indicate a significant difference between breast tumor and normal breast as well as diseased lymph nodes and normal lymph nodes with p values inset in the graphs. (B) CAIX and CAXII protein staining patterns in normal breast, lymph node metastases and adjacent normal tissue. For representative images of CAIX (a) and CAXII (b), staining is localized to the normal breast epithelium. Positive membrane staining in lymph node tumor metastasis (black arrows) and adjacent unaffected lymph nodes for CAIX (c) and CAXII (d) (scale = 25 μm).

Figure 2

Expression of CAIX and CAXII in MDA-mb-231 cells, and verification of specific binding of CA9Ab-680 and CAX12Ab-680. Confocal micrographs of cells incubated with the nuclear marker DAPI (blue), the plasma- and cytoplasmic-membrane marker, WGA (green) and CA9 and CA12Ab-680 (red). (A) CA9Ab-680 (red) stained MDA-mb-231 cells that constitutively express CAIX, but not MCF-7 cells which do not express CAIX in normoxic conditions. (B) CA12Ab-680 (red) stained MDA-mb-231 cells engineered to express CAXII (up). CA12Ab-680 stained parental MDA-mb-231 cells that do not express CAXII. Note that CA12Ab-680 does not stain these cells (bottom). Please note that merged images show colocalization of agent (red) with membrane marker (green) indicating accumulation of agent on the cell-surface.

Figure 3

In vivo and ex vivo selectivity of CA9Ab-680 and CA12Ab-680. (A) In vivo fluorescence image of mouse bearing xenograft tumors, 24 hours post intravenous injection of agents: MDA-mb-231 cells that constitutively express CAIX were used to form the positive tumor (left panel). A blocking experiment was performed using pre-injection of 150 μg unlabeled CAIX mAb, followed by injection of 50 μg CA9Ab-680 to determine specificity of the CAIX specific probe (middle panel). The image was acquired 24 hours after injection with labeled probe. MDA-mb-231/CAXII cells were used to form the positive (+) tumor and parental cells were used as the negative (-) tumor (right panel). Tumors are indicated by orange arrows. The agents have cleared from the entire mouse, including the negative tumor at this time-point. (B) Confocal microscopy of negative and positive tumor sections stained with CA9 and CA12Ab-680 (red), membrane marker (green) and nuclear (blue) counterstain. Note the negative tumor is CAXII negative and was stained with CA12Ab-680. (C) Ex vivo images of tumors with corresponding histology (CAIX and CAXII IHC staining) of positive and negative tumors.

Figure 4

Pharmacodynamics and biodistribution of CA9Ab-680 and CA12Ab-680. (A) Pharmacodynamics of agents in positive and negative tumors. Note that the peak signal in the positive CAIX and CAXII tumors is 24-hours post-injection, and that the agents are nearly cleared after 8 days. Data represent mean ± s.d. (B) Biodistribution of agents in the CAIX and CAXII tumors and organs, 24 hours and 48 hours post-injection shows predominant positive tumor localization and clearing from the organs by 48 hours post-injection. The values were normalized as percentage of the highest signal.

Figure 5

Representative images of spontaneous metastases into axillary lymph node of MDA-mb-231/Luc (A,B) and MDA-mb-231/Luc/CA12 (C,D) cells and the ability of the agents to detect the tumor cells in ALN. (A,C) Bioluminescence images of luciferase activity in the primary tumor (yellow arrow) and axillary lymph node (red arrow) of MDA-mb-231/Luc (A) and MDA-mb-231/Luc/CA12 (C) xenografts. Fluorescence surface radiance from primary tumor, injection site and axillary lymph node 24 hours post peritumoral injection of CA9Ab-680 (B) and CA12Ab-680 (D).

Figure 6

Sensitivity of CA9Ab-680 and CA12Ab-680 for detection of tumor cells in ALN. A range of positive cells (MDA-mb-231 for CAIX and MDA-mb-231 expressing CAXII for CAXII) were injected into ALN using ultrasound image guidance. (A) Bioluminescence activity quantified for a range of injected cell numbers. Insets show signal for mice injected with the lowest cell number. (B) Agent associated fluorescence for a range of cell numbers, 24 hours post-injection. Insets show fluorescence for mice injected with minimum detectable cells by agents (1000 cells for CA9Ab-680 and 500 cells for CA12Ab-680). All data represent mean ± s.d. of pixel values within the ROIs.