Quantum dots in axillary lymph node mapping: Biodistribution study in healthy mice

Breast cancer is the first cause of cancer death with a 100% rising in incidence from 1980 to 2000 and 35% growth in mortality [1]. Axillary lymph node dissection (ALND) is the standard surgical treatment in breast cancer patients. It consists in whole lymphatic axillary chain removal with the aim to reveal presence of metastases by histology analyses. However, this procedure leads to numerous morbidity problems such as pain, lymphocel, restriction of arm motion, chronic lymphedema, paresthesia, and, as a consequence, deterioration of life quality. For these reasons, an alternative treatment consisting in the sentinel lymph node (SLN) biopsy has been met with enthusiasm by clinicians. The SLN is the first node in the lymphatic basin into which drains the primary tumour [2]. As a result, axillary lymph node status is the most important prognostic factor and a determinant in the choice of therapy [3]. SLN biopsy consists in ^99mTc-radiolabeled sulfur colloid injection the day before surgery in peritumoral or periareolar area [4] followed by lymphoscintigraphy to identify radioactive lymph node(s). During the surgery, a blue dye is injected in the same way and SLN is thus identified as blue and hot lymph node, one of these criterions only is not sufficient for a successful SLN identification [5]. Despite the significant improvement of SLN biopsy over ALDN, the problems related to potential radiation hazards along with an extra cost of the radioisotope, to allergic reactions after blue dye injection and the overall procedure duration are limiting factors for this technique [6]. The ideal method should be sensitive, accurate, rapid, non invasive, non radioactive, and potentially usable in a laparoscopic setting for gastrointestinal tumors [7]. None of the current methods fulfills all of these criteria. Lymphatic mapping with blue dye results in a high rate of false-positive nodes because the small dye particles can readily diffuse through the true SLN and traverse multiple nodes [8]. Additionally, blue dye has poor tissue contrast and is difficult to detect in deep, dark anatomical regions such as the abdomen. Although the use of radioisotope tracers has improved the detection rate and accuracy of SLN mapping, the high radioactivity of the primary injection site can interfere with intraoperative in vivo detection of nearby nodes [3]. Additionally, if radioisotopes are injected during surgery, the time period required for the tracer to migrate to the SLN may delay the operative procedure. Conversely, preoperative injection of the radiocolloid tracer often necessitates anywhere from 2 to 16 hours before surgery. This is difficult, if not impossible, to accomplish and is applicable only in endoscopically accessible regions of the colon and stomach [7].

Quantum dots (QDs) are fluorescent inorganic nanometer sized crystals with remarkable unique optical and electronic properties. They are composed of a fluorescent core of semiconductor heavy metals. In order to limit their potentially toxic effects due to the release of Cd²⁺ or Se^2- ions in case of oxidation of the core (by ultraviolet (UV) light or oxygen) and to passivate the defects on its surface, this core is encapsulated in a shell of organic material, polymeric or lipid-based layers then coat the shell. They have a good quantum yield and absorption cross-section such that single QD fluorescence imaging is possible in cells. QDs have been already applied to visualize SLN in gastrointestinal tract [7], pleural space [9], lung area [10], oesophageal area [11], skin [12] and axilla [13] using infrared light. Recently, red and infra-red emitting QDs were imaged in inguinal and axillary lymph nodes in tumor bearing mice already 3 min after local injection [14]. Major difficulty in using QDs in vivo is their unknown potential toxicity. However, it has been demonstrated that the injection of 2.10⁹ QDs/cell in xenopus embryos unaltered their phenotype and health [15] and that QDs triggered apoptosis via reactive oxygen species (ROS) production only when they are not coated with ZnS [16, 17]. QDs toxicity study is ultimately related to the investigation of their biodistribution in pre-clinical models. Currently, only few detailed works on QDs biodistribution in rodents have been realized [18–20].

The primary objective of this work was to detect and quantify at different times after the injection the selective accumulation of carboxyl coated QDs in axillary draining lymph nodes. Secondly, we assessed the biodistribution of QDs in different tissues by fluorescence microscopy and chemical extraction procedures. The rationale for the choice of carboxyl coating was based on the observation that negative charge confine to QDs certain selectivity to reticuloendothelial system (RES) [21] along with the excellent retention in lymph nodes [22]. Red emitting 655 nm QDs were considered in this study because of their better fluorescence quantum yield compared to that of IR QDs (fluorescence quantum yield) allowing better quantification with regard to detection level of spectroscopic tools.

In vivo biodistribution studies of QDs are sparse and deal mostly with intravenously (IV) injected QDs. The general consensus is that QDs distribution depends on their physicochemical properties and in particular, size and coating. Tracer particle size has an important effect on migration time in SLN mapping. Particles < 5 nm diffuse into blood, those between 5 to 10 nm can migrate through nodal tissue and result in false-positive results, and those > 1000 nm largely remain at the injection site. Consequently, we choose QDs with a diameter of 16 nm, since a size range between 15 and 20 nm is optimal to enable the QDs to travel through the lymph channels and be trapped by SLN [9, 12]. Concerning coating, we choose carboxyl coated QDs. The carboxyl groups present the property to be negatively charged and therefore display rapid uptake in lymphatics and excellent retention in lymph nodes [12, 21, 24].

To our knowledge, a single biodistribution study was carried out after non-IV administration of QDs. Gopee et al. [19] investigated the distribution in healthy mice of 48 pmol intradermally (ID) injected QDs with a diameter of 37 nm and PEG coating. The authors did not find any QDs in lymph nodes until 12 h after injection, whereas we could detect them already 5 minutes following the injection (Figure 1). An early detection (3 min) of QDs traveling toward inguinal nodes was observed after intratumoral injection of red emitting QDs at concentrations similar to ours [14]. However this observation was not followed by biodistribution investigations. QDs Fluorescence increases with time to reach a maximum at 60 min after the injection. The calculated percentages of the injected dose (2.42% at 60 min and 1.24% at 24 h) (Table 1) were comparable with those of Kim et al. [13] and Gopee et al. [19] who found 2–4% and 1.07% respectively in the SLN. We observed a 4 fold decrease of QDs concentration in ALN from 60 min to 24 h post-injection (Figure 1). We also noted a strong fluorescence of QDs at the injection site. This observation is consistent with the study of Gopee and co-workers [19] who found 60% of cadmium retention at the injection site. We further undertook the study of the biodistribution of QDs in the organs. No fluorescence peak was observed after chemical extraction of Qdot^® 655 ITK™ carboxyl in liver, kidneys, spleen, lungs, heart, intestine, brain, bladder, and skin (Figure 3). This points out that we are below the calculated concentration limit (Table 2), even if the latter is as low as around 2.5 pmol/g of tissue. The lower detection threshold observed for pigmented organs such as liver and heart was expected (Table 2) since the fluorescence detection method employed in our study is very sensitive to the absorption properties of the sample. In contrast to our results, the previous QDs biodistribution studies all evidenced the presence of QDs in the liver and also to a minor extent in spleen and kidneys [18, 19]. The plausible explanation could be provided by the difference in the site of injection. Gopee et al. [19] injected QDs ID in the dorsal flank and found 24 h after injection 6.40% of the initial dose in liver and a small amount in kidneys (0.51%) and spleen (0.18%). We have administrated QDs subcutaneously in the paw. ID injection of QDs in the flank drains differently from the SC injected QDs [25] because of the anatomical differences in lymphatic network of the two regions. Likewise, when intradermal injection is performed on the back, direct lymphatic drainage to retroperitoneal and paravertebral nodes can occur before the injected product is distributed throughout the body [26]. When injection is performed in the paw, the first lymph node that receives direct lymphatic drainage is always in the axillary area. QDs are unlikely to be excreted from the body, neither by urine nor by feces in the time span of 48 h. This observation is in agreement with previous work [18] demonstrating the absence of QDs in urine and feces during ten days. Theses results are explained by the dramatic reduction of renal filtration for QDs with a hydrodynamic diameter higher than 3.5 nm and with negative charges [27]. Taken as a whole, the decline in QDs concentration 24 h post-injection could not be attributed to the elimination and/or migration of QDs to different organs. Assuming very strong retention in SLN, we hypothesize that QDs may diffuse in the second lymph node of the chain. This supposition is consistent with the fluorescence distribution pattern of QDs in SLN (Figure 2). Indeed QDs fluorescence is extended till medulla sinus, thus rending possible the migration to the distal lymph node through the lymph channels. The interaction of QDs with serum should also be addressed. We did not perform studies on the interaction of carboxyl QDs with serum proteins, however, as it was shown recently anionic or cationic charged QDs were associated with an increase of the hydrodynamic diameter after incubation with serum [20]. Moreover, the intensity of fluorescence of QDs increases in the presence of BSA [28] or hemoglobin [29].

Spectrofluorometry using an optical fiber is a non-invasive procedure for the detection of fluorescent tracers. The results of our study demonstrate that despite the incapacity to detect QDs directly through the skin, LIF measurements allow the detection of QDs in ALN after skin incision with a profile similar to chemical extraction (Figure 4). However, non-invasive measurements are frequently suffering from significant variations of measurements, related to light diffusion and scattering. Further, actually, our work is a preliminary step to use of NIR emitting QDs which we expect the detection of QDs through the skin without any incision. In the future, the non-invasive measurements could be achieved by modulating an excitation wavelength and using NIR emitting QDs. The excitation wavelength in our study was set to 410 nm. Using an excitation at larger wavelengths should result in less skin absorption and scattering, and should make deep tissue imaging possible. Detection of QDs with optical fiber spectroscopy opens perspectives for the utilization of this technique in operation theatre, since it can provide a rapid and reliable QDs detection immediately after SLN excision. A limitation to the clinical use of QDs for SLN mapping is their potential toxicity [30]. However, because most of the injected dose is eliminated by removal of the injection site and nodal tissue, the eventual toxicity may be negligible. Furthermore, to avoid toxicity caused by biodegradation of QDs and exposition of core metal ions, QDs with non-heavy metal cores (such as indium) would be developed along with ameliorations of surface chemistry [31].

In this study we demonstrated that 655 nm emitting QDs with carboxyl coating allow the visualization of axillary draining lymph node (ALN) after subcutaneous injection of a very weak quantity of QDs (20 pmol). Unlike other tracers, QDs rapidly localize to and remain trapped in the SLN. A strong fluorescence signal is obtained within a few minutes lasting at least 24 h as measured with optical fiber spectrofluorometry, a minimally invasive technique allowing rapid pre-operative complete procedure. Fluorescence microscopy and chemical extraction confirmed highly selective QDs accumulation in the ALN corresponding to 2.42% of the injected dose at 60 min after the injection; the major part remaining at injection site. This rapid and selective accumulation of QDs in ALN together with the eventual non-invasive fluorescence detection offers an exciting opportunity to track lymphatic flow in real time and guide their nodal dissection after one simple intraoperative injection.