Evaluation of a targeted anti-αvβ3 integrin near-infrared fluorescent dye for fluorescence-guided resection of naturally occurring soft tissue sarcomas in dogs

Abstract

Purpose Complete resection is a key prognostic factor for survival in patients with soft tissue sarcoma (STS), in humas and companion animals alike. Fluorescence-guided surgery could improve resection accuracy. As dogs are frequently affected by STS, they serve as a model to test an anti-αvβ3 integrin targeting near-infrared fluorescent (NIRF) dye (AngiostampTM800) for fluorescence-guided surgery in STS to evaluate its safety and feasibility in dogs, and if it translates into a clinically relevant benefit compared to the standard of care with regards to completeness of surgery and local recurrence. Furthermore, we aimed to correlate target expression and NIRF-signal intensity. Methods Twenty dogs with STS were randomly allocated to either receive Angiostamp™ (NIRF group) or physiologic saline (control group) preoperatively. The researchers were blinded for treatment, and resections were adapted based on the NIRF-signal, if needed. Margin status was histologically determined at the 1 and 3 cm margin. The tumor-to-background ratio was measured in native tissue biopsies and formalin-fixed tissue. The fluorescent area was compared to the corresponding tumor areas as confirmed by histology using the Dice coefficient. Target expression was quantified by immunohistochemistry and correlated to NIRF-signal ratios. Results A fluorescent signal was detected in all 10 tumors of the NIRF group, with a tumor-to-background ratio of 7.4 ± 5.8 in native biopsies and 13.5 ± 10.9 in formalin-fixed tissue. In the NIRF group, resection margins were adapted in 5/10 cases, leading to complete resection and preventing R1 in four of these cases. In the NIRF and control group 9/10 and 8/10 resections were R0, with one local recurrence in each group and one sarcoma-related death in the NIRF group. The NIRF-signal correlated with the histologically confirmed tumor area (Dice coefficient 0.75 ± 0.17). Target expression was higher in tumor compared to peritumoral tissue (p < 0.0003) and showed a moderate correlation with the NIRF-signal (r = 0.6516, p < 0.0001). Conclusion Fluorescence-guided surgery using Angiostamp™ can pinpoint residual disease in the tumor bed and contributes to an improved resection accuracy in canine STS.