Introdução: O algoritmo terapêutico do carcinoma hepatocelular (CHC) é complexo e tem na ressecção cirúrgica um dos seus principais alicerces. Contudo, a recidiva pós-operatória ainda é alta, variando de 50 a 80% em 5 anos. Por esta razão, a busca por fatores prognósticos que possam refinar o entendimento da biologia e comportamento tumorais tem aumentado. Estudos recentes têm demonstrado a relação entre a resposta inflamatória e o prognóstico oncológico de diversas neoplasias sólidas gastrointestinais. Entretanto, o impacto prognóstico dos marcadores inflamatórios em pacientes submetidos à hepatectomia por CHC ainda é pouco conhecido. Os estudos existentes têm mostrado resultados promissores, no entanto, são, em sua maioria, oriundos de centros orientais, com características tumorais e epidemiológicas distintas das encontradas em centros ocidentais. Objetivo: O objetivo primário é avaliar o impacto prognóstico dos principais marcadores inflamatórios pré-operatórios relação neutrófilo-linfócito (RNL), relação plaqueta-linfócito (RPL) e relação monócito-linfócito (RML) no desfecho oncológico de pacientes submetidos à hepatectomia curativa por CHC. O objetivo secundário é estudar o impacto destes marcadores em subgrupos de pacientes com CHC quanto ao seu tamanho: < 5 cm, entre 5 e 10 cm e > 10 cm. Método: Foram estudados, tendo como base um banco de dados prospectivo, pacientes operados de janeiro de 2007 a dezembro de 2018. Os critérios de exclusão foram: doença extra-hepática, ressecção R1/R2 e tratamentos locorregionais prévios. Os melhores cut-offs para a RNL, RPL e RML foram calculados a partir da construção de curvas ROC (Receiver Operator Curve) e do índice de Youden, dividindo-se os pacientes em 2 grupos, acima e abaixo do cut-off calculado. As sobrevidas, global (SG) e livre de doença (SLD), foram estimadas por meio do método de Kaplan-Meier e comparadas pelo teste de log-rank. O método de Cox foi utilizado para identificar fatores independentes associados à SG e SLD. O nível de significância foi 5%. Resultado: Foram incluídos 161 pacientes. Com base na curva ROC, os melhores pontos de corte relacionados à mortalidade foram: RNL (1,715), RPL (115,05) e RML (1,750). Para recidiva, os pontos de corte foram: RNL (2,475), RPL (100,25) e RML (2,680). A RNL elevada ( > 1,715) associou-se à pior SG (p=0,018). A RNL ( > 2,475; p=0,047) e RPL ( > 100,25; p=0,028) elevadas associaram-se à pior SLD. Em pacientes com CHC < 5 cm, a RML ( > 1,715) foi associada à menor SG (p=0,047). Na análise multivariada, a hipertensão portal (HR=7,04; IC95% 2,40-20,66; p < 0,001), AST > 50 U/dL (HR=3,06; IC95% 1,10-8,47; p=0,032) e tempo em UTI > 3 dias (HR=5,04; IC95% 1,75-14,49; p=0,003) foram fatores associados à pior SG. Quanto à SLD, AST > 50 U/dL (HR=3,32; IC95% 1,60-6,91; p=0,001), invasão vascular (HR=2,36; IC95% 1,13-4,93 p=0,022) e RPL elevada (HR=3,03 IC95% 1,50-6,12; p=0,002) foram independentemente associados à maior recidiva. Conclusão: RNL elevada foi associada à menor SG e SLD. A RPL elevada foi preditora independente de pior SLD. A RML elevada foi preditora de menor SG em pacientes com CHC < 5 cm

Background: The therapeutic decision algorithm for hepatocellular carcinoma (HCC) is complex and has surgical resection as the mainstay of treatment. However, the post-operative recurrence of hepatocellular carcinoma is still high, reaching 5080% in five years. For this reason, the search for prognostic markers that may help shed more light on tumor biology has been growing. For instance, the relationship between the inflammatory response and the oncologic prognosis of solid gastrointestinal tumors has recently been accessed. The available studies, reported from a majority derived of Eastern centers, have shown promising outcomes involving distinct tumor and epidemiologic features, in comparison to their Western counterparts. Despite this, the impact of inflammatory markers in oncologic outcomes of patients who underwent hepatectomy for HCC is still poorly known in general. Aim: The primary endpoint was to evaluate the prognostic impact of the main pre-operative inflammatory markersneutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and monocyte-to-lymphocyte ratio (MLR)in the long-term outcomes of patients who underwent curative hepatectomy for HCC. The secondary endpoint was to evaluate the prognostic impact of these markers in subgroups of patients according to tumor size: < 5 cm, 510 cm, and > 10 cm. Method: A prospective database was utilized to study patients who underwent HCC resection from January 2007 to June 2018. The exclusion criteria were as follows: extra-hepatic disease; R1/R2 resection, and preoperative locoregional treatments. The best cut-off values for NLR, PLR and MLR were calculated through Receiver Operator Characteristic (ROC) curve analysis using the Youdens index. Thereafter, patients were divided into two groups: those below and above the calculated cut-offs. Overall survival (OS) and disease-free survival (DFS) were estimated using the Kaplan-Meier method and compared by the log-rank test. The Cox method was used to identify independent predictors associated with OS and DFS. The significance level used was set to 5%. Results: A total of 161 patients were enrolled. Based on the ROC curve analysis, the best cut-offs regarding mortality were the following: NLR (1.715), PLR (115.05), and MLR (1.750). For recurrence, the best cut-offs were NLR (2.745), PLR (100.25), and MLR (2.680). A high NLR (1.715) was associated with worse OS (p=0.018). High NLR (>2,475; p=0.047) and PLR ( > 100.25; p=0.028) were associated with worse DFS. In HCC < 5 cm, the MLR ( < 1.715) was associated with worse OS (P=0.047). In multivariate analysis, portal hypertension (hazard ratio [HR]=7.04; 95% confidence interval [CI] 2.40-20.67; p < 0.001), aspartate aminotransferase (AST) > 50 U/dL (HR=3.06; 95% CI 1.10-8.47; p=0.032), and ICU stay > 3 days (HR=5.04; 95% CI 1.75-14.49; p=0.003) were factors associated with worse OS. For DFS, AST > 50 U/dL (HR=3.32; 95% CI 1.60-6.91; p=0,001), vascular invasion (HR=2.36; 95% CI 1.13-4.93 p=0.022), and elevated PLR (HR=3.03; 95% CI 1.50-6.12; p=0.002) were independent predictors for higher recurrence. Conclusions: An elevated NLR was related to worse OS and DFS. An elevated PLR was an independent predictor of worse DFS. An elevated MLR was a predictor of worse OS in patients with HCC < 5 cm