DIAGNOSTIC EVALUATION OF CLINICAL COMPLETE RESPONSE BY PROBE-BASED CONFOCAL ENDOMICROSCOPY AFTER NEOADJUVANT CHEMORADIATION FOR LOCALLY ADVANCED RECTAL ADENOCARCINOMA
Adriana V. Safatle-Ribeiro*2, Carlos F. Marques2, Clelma Batista Pires2, Lívia Arraes2, Luciana Meirelles2, Elisa R. Baba2, Andressa A. Machado2, Martin A. Coronel2, Gustavo R. Lima2, Iatagan Josino2, Diogo Araujo2, Evelise Pelegrinelli-Zaidan2, Carla C. Gusmon2, Marcelo S. de Lima2, Caterina Pennacchi2, Fabio S. Kawaguti2, Ricardo S. Uemura2, Bruno da Costa Martins2, Gustavo A. de Paulo2, Luciano Lenz2, Guilherme C. Cotti2, Caio Sergio R. Nahas2, Sergio C. Nahas2,1, Ulysses Ribeiro1,2, Fauze Maluf-Filho2
1Gastroenterology, University of Sao Paulo , Sao Paulo, SP, Brazil; 2Cancer Institute of the University of Sao Paulo (ICESP), Sao Paulo, Sao Paulo, Brazil
Neoadjuvant chemoradiotherapy (nCRxt) followed by surgery represents an optimal approach for medium and distal advanced rectal adenocarcinoma. Patients with complete clinical response (cCR) may be followed closely without immediate surgery (watch and wait strategy). Magnetic resonance imaging (MRI) and endoscopy have low accuracy for the diagnosis of cCR. Probe-based confocal endomicroscopy (pCLE) is a real time in vivo method that allow acquisition of optical biopsies with 1000 times magnification, evaluating both cellular and vascular patterns. Aim: To evaluate the role of probe-based confocal endomicroscopy (pCLE) in the diagnosis of cCR after nCRxt for advanced rectal adenocarcinoma. Methods: We evaluated 47 patients with locally advanced rectal adenocarcinoma (T3 - T4, or N+) who underwent nCRxt (5-fluoroucacil, 5040 cGy). pCLE was performed post nCRxt, and guided tissue biopsies subsequently. Images of pCLE post nCRxt were recorded and blinded compared to the histology of the surgical specimens. Vascular score comprises evaluation of vessel enlargement, vessel tortuosity, leakage, and defective flux. During pCLE, additional parameters involving the vessels (endothelium thickening and vascular looping enlargement); and the stroma (fibrotic and chicken skin) were also added to the evaluation. Epithelial features, including dark and irregular cells, presence of budding, back-to-back glands, cribriform pattern, and increased vase/crypt ratio were also considered. Results: Twenty-seven (57.5%) patients were men, and the mean age was: 62.8 years, varying from 35 to 82 years. Twenty-four patients had tumor lesions located at medium rectum and 23 patients at distal rectum. Thirty-seven had partial response confirmed by pCLE. Ten patients (21.3%) had good endoscopic response, and presented small ulcer (n=5) or residual scar (n=5). MRI correctly diagnosed 3/10 of these patients and pCLE correctly diagnosed 8/10 patients. pCLE and MRI misdiagnosed a patient who was considered positive, however pathology report showed mucin areas without neoplastic cells. pCLE sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 100%, 66.6%, 95.3%, 100% and 95.7%, respectively. After nCRxt, enlarged and tortuous vessels besides the presence of budding, back-to-back glands, cribriform pattern and increased vessel/crypt ratio were essential parameters to differentiate malignancy from post-radiation alterations and inflammation. pTNM stages of this subgroup were: 4 ypT0 ypN0, 1 ypT0 ypN1, 1 ypT1 ypN0, 2 ypT2ypN0, and 2 cT0cN0. Conclusions: 1. pCLE epithelial and vascular features may improve the diagnosis of cCR, and may alter patient management; 2. pCLE might be valuable for identifying patients with advanced rectal cancer who will benefit from watch and wait policy, avoiding immediate surgical treatment.
Back to 2020 Abstracts