Clinical and diagnostic implications of sentinel lymph node detection in canine mast cell tumours

The studies focused on canine mast cell tumour (MCT) have exponentially risen in the last ten years, making the MCT a “hot topic” in veterinary oncology and an object of debate in the clinical and research field. The sentinel lymph node (SLN) is the first node within the lymphatic basin that directly drains the primary tumour. The previously published studies show that the non-concordance rate between regional (RLN) and SLN is largely variable, ranging from 22.3% to 63%. In this dissertation, we evaluated the agreement rate between SLN and RLN using indirect computed tomography lymphangiography (ICTL) and intraoperative methylene-blued dye (MB) peritumoral injection. We also described the agreement between the two techniques and focused on all the possible reasons for failure of identification of SLN. Sixty-one client-owned dogs with 61 MCTs were prospectively enrolled. RLN was predicted based on the lymphosomes concept. ICTL had a SLN detection rate of 80% (49/61), while in 20% of cases, neither the SLN nor the lymphatic duct could be identified. We observed that ICTL failed to identify SLN when the MCT had a diameter greater than 6 cm. Moreover, even if not statistically significant, we reported a tendency to fail in case of the presence of the tomographic sign of “peritumoral halo”. The SLN-RLN agreement was complete in 64% (n=39) and partial in 10% (n=6) cases. Intraoperative MB was peritumorally injected in 47 dogs. In 75% (n=35) the SLN-RLN concordance was total, in 19% (n=9) was partial and in 6% (n=3) was absent. In 68% (32/47) the agreement between ICTL and MB was total and partial in 3 cases (6.4%); in the remaining 12 cases (25.5%) there was a complete disagreement. Secondly, we described the incidence rate and the severity of the complications following peripheral lymphadenectomy in dogs, comparing three different guiding techniques: intraoperative γ-probing and methylene blue, MB only, and unguided lymphadenectomies. Client-owned tumour-bearing dogs undergoing both pre-operative SLN mapping and excision of peripheral SLNs were included. A total of 201 peripherally lymphadenectomies from 163 dogs from 5 institutions were enrolled. Seventy-two (59%) lymphadenectomies were performed without intraoperative guidance, 49 (41%) with MB, and 80 (40%) with γ-probe and MB. The overall incidence of complications reported was 7.5% (15/201), and 80% (12/15) of them were mild. In the logistic regression model, the use of intraoperative guidance and none of the factors considered significantly influenced the complication rate. The most frequent postoperative event was seroma (2.5%). Using the decision tree model, mandibular and retropharyngeal lymphadenectomy significantly affected the complication rate, especially when surgery lasted more than 21.5 minutes. Lastly, we evaluated whether the Ki67 index has a predictive value in a homogeneous cohort of Patnaik grade 2/Kiupel low-grade (G2/LG) cMCTs with HN2 LNs. The medical databases of three institutions were retrospectively searched for dogs undergoing surgical treatment for cMCT and LN extirpation, with a histological diagnosis of G2/LG with HN2 LNs. A total of 39 cases were identified. None of these dogs developed local and nodal relapse or metastatic distant disease. At the end of the study, 32 (82%) dogs were alive, 7 (18%) dogs were dead from unrelated causes and 4 (10.2%) dogs were lost to follow-up. Considering the strict inclusion criteria, dogs affected by G2/LG with HN2 LNs treated with surgery alone may have a good oncologic outcome; the Ki67 index does not have a prognostic impact. In conclusion, sentinel lymph node mapping and excision play an important role in the diagnosis and therapy of canine MCTs, the technique used for the detection is crucial and should be chosen based on the clinical presentation of the tumour to avoid failures. Anyway, the lymphadenectomy per se is usually a rather safe procedure, rarely associated with severe complications.