Quantitative Primary Tumor Indocyanine Green Measurements Predict Osteosarcoma Metastatic Lung Burden in a Mouse Model

Abstract

Background: Current preclinical osteosarcoma (OS) models largely focus on quantifying primary tumor burden. However, most fatalities from OS are caused by metastatic disease. The quantification of metastatic OS currently relies on CT, which is limited by motion artifact, requires intravenous contrast, and can be technically demanding in the preclinical setting. We describe the ability for indocyanine green (ICG) fluorescence angiography to quantify primary and metastatic OS in a previously validated orthotopic, immunocompetent mouse model.

Questions/purposes: (1) Can near-infrared ICG fluorescence be used to attach a comparable, quantitative value to the primary OS tumor in our experimental mouse model? (2) Will primary tumor fluorescence differ in mice that go on to develop metastatic lung disease? (3) Does primary tumor fluorescence correlate with tumor volume measured with CT?

Methods: Six groups of 4- to 6-week-old immunocompetent Balb/c mice (n = 6 per group) received paraphyseal injections into their left hindlimb proximal tibia consisting of variable numbers of K7M2 mouse OS cells. A hindlimb transfemoral amputation was performed 4 weeks after injection with euthanasia and lung extraction performed 10 weeks after injection. Histologic examination of lung and primary tumor specimens confirmed ICG localization only within the tumor bed.

Results: Mice with visible or palpable tumor growth had greater hindlimb fluorescence (3.5 ± 2.3 arbitrary perfusion units [APU], defined as the fluorescence pixel return normalized by the detector) compared with those with a negative examination (0.71 ± 0.38 APU, -2.7 ± 0.5 mean difference, 95% confidence interval -3.7 to -1.8, p < 0.001). A strong linear trend (r = 0.81, p < 0.01) was observed between primary tumor and lung fluorescence, suggesting that quantitative ICG tumor fluorescence is directly related to eventual metastatic burden. We did not find a correlation (r = 0.04, p = 0.45) between normalized primary tumor fluorescence and CT volumetric measurements.

Conclusions: We demonstrate a novel methodology for quantifying primary and metastatic OS in a previously validated, immunocompetent, orthotopic mouse model. Quantitative fluorescence of the primary tumor with ICG angiography is linearly related to metastatic burden, a relationship that does not exist with respect to clinical tumor size. This highlights the potential utility of ICG near-infrared fluorescence imaging as a valuable preclinical proof-of-concept modality. Future experimental work will use this model to evaluate the efficacy of novel OS small molecule inhibitors.

Clinical relevance: Given the histologic localization of ICG to only the tumor bed, we envision the clinical use of ICG angiography as an intraoperative margin and tumor detector. Such a tool may be used as an alternative to intraoperative histology to confirm negative primary tumor margins or as a valuable tool for debulking surgeries in vulnerable anatomic locations. Because we have demonstrated the successful preservation of ICG in frozen tumor samples, future work will focus on blinded validation of this modality in observational human trials, comparing the ICG fluorescence of harvested tissue samples with fresh frozen pathology.

Fig. 1

A flowchart shows the experimental protocol. K7M2 OS cells were injected paraphyseally into each animal’s left hindlimb. Four weeks after OS injection, sharp transfemoral amputations were performed of the left hindlimb to remove the primary tumor. Euthanasia with lung harvesting was performed 10 weeks after OS injection. Twenty-four hours before both amputation and euthanasia, 30 μL of 2.5 mg/mL ICG was injected retroorbitally. Postwashout ICG fluorescence measurements were performed immediately before amputation or on the lungs ex vivo.

Fig. 2 A-B

(A) Hindlimb fluorescence shows K7M2 OS cells injected into the left hindlimb of a 4- to 6-week-old female Balb/c mouse. Red represents a high fluorescent signal, whereas blue denotes poor/no signal above the background. (B) Representation of our standardized measurement of primary tumor fluorescence using three independent points on the hindlimb. Normalization for between-animal comparisons was performed by dividing the average hindlimb perfusion with the maximum signal measured on the animal’s tail.

Fig. 3 A-B

(A) Metastatic OS in the lungs is shown with increased ICG fluorescent signal. In contrast, lungs without metastatic disease (B) showed minimal ICG signal above baseline.

Fig. 4 A-B

Histologic fluorescence shows ICG on flash frozen samples of the (A) primary tumor (green signal = ICG) and (B) lungs. The yellow arrow highlights areas of metastatic osteosarcoma in the lungs, which fluoresce green because of ICG localization. Normal tissue in the lungs and the lower extremity did not fluoresce.

Fig. 5

Hindlimb fluorescence 4 weeks after K7M2 OS injection was greater in animals that developed primary tumors compared with those that did not.

Fig. 6

The primary tumor fluorescence 4 weeks after K7M2 OS injection when stratified by cell injection group. The 1 x 10⁶ OS cell group had greater fluorescence than all other animal groups (10.7 ± 8.8, 9.5 mean difference, 91% CI 1.2-17.9, p < 0.05). Asterisks indicate statistical significance (p < .05) compared with controls.

Fig. 7 A-B

(A) Ex vivo lung specimen showing clinically obvious metastatic OS. (B) ICG fluorescence of this same specimen. Red = high ICG signal; blue = low/no signal above the background.

Fig. 8

A linear relationship was observed between hindlimb and lung fluorescence (r² = 0.81, p < 0.01).

Fig. 9

No relationship was observed between hindlimb fluorescence and tumor volume as measured with CT (r² = 0.04, p = 0.45).