A carregar...
Publicação
Immune-related serious adverse events in immune checkpoint inhibitors in cancer patients : systematic review and network meta-analysis
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 8.72 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Objetivo Estimar a frequência de eventos adversos imunomediados em doentes com cancro que receberam inibidores de checkpoint imunitário (ICI). Métodos A pesquisa foi realizada nas bases de dados CENTRAL, MEDLINE, Embase, e clinicaltrials. gov, até fevereiro de 2021. Foram incluídos ensaios clínicos aleatorizados e controlados, com qualquer ICI como intervenção, para qualquer tipo de cancro. O objetivo primário foi estimar a incidência global de eventos adversos imunomediados graves. A revisão foi realizada de forma independente nas várias fases, nomeadamente de seleção, avaliação do risco de viés e extração de dados. Realizou-se meta análise em rede Bayesiana, com comparação entre classes de ICI e entre ICI em monoterapia. Realizou-se ranking das modalidades terapêuticas (usando a superfície da curva cumulativa de ranking [SUCRA]) e estimou-se a frequência mediana para cada outcome. Resultados Foram incluídos 96 ensaios com um total de 52,811 participantes, publicados entre 2011 e 2021. O tamanho amostral mediano foi de 559, sendo a idade mediana dos participantes 62 anos com 37% do sexo feminino. Apenas 35.4% dos ensaios utilizaram ocultação dupla e 82% tiveram elevado risco de viés. O cancro do pulmão de não pequenas células (28 ensaios, 17,014 participantes) e o melanoma (15 ensaios, 8,808 doentes) foram os mais frequentemente analisados. Pembrolizumab (14 ensaios, 9,322 participantes) e nivolumab (14 ensaios, 5,779 participantes) foram as principais intervenções. Para todos os outcomes, a classe anti-CTLA4 e a combinação entre ICIs foram as modalidades com maior probabilidade global de eventos imunomediados sérios (SUCRA de 7% e 10%, respetivamente), tendo ambas, face à terapêutica convencional, odds aumentadas destes eventos (OR 24.41, 95% CrI 11.11 a 58.04 e 22.46, 95% CrI 11.86 to 44.91, respetivamente).O ranking dos ICI individuais foi estimado, sendo o ipilimumab o pior classificado (SUCRA 14%), com taxa absoluta de eventos de 848 por 10,000 doentes (intervalo de confiança a 95% de 252 a 2,610) e sendo o atezolizumab o melhor classificado (SUCRA 82%), com taxa absoluta de eventos de 119 por 10,000 doentes (intervalo de confiança a 95% de 54 a 279). Conclusão O risco global de desenvolver eventos adversos graves imunomediados difere entre cada ICI e para cada evento específico considerado. A diferente magnitude do aumento do risco de cada evento com cada modalidade de tratamento poderá ser clinicamente relevante, influenciando o manejo individual dos doentes com cancro.
Objective To estimate the frequency of immune-related serious adverse events (irSAEs) in cancer patients receiving immune checkpoint inhibitors (ICI). Methods We searched CENTRAL, MEDLINE, Embase, and clinicaltrials.gov for randomized controlled trials from inception to February 2021. We included randomized controlled trials investigating any ICI for any type of cancer. The primary outcome was overall irSAEs, and secondary outcomes included specific irSAEs of interest. Screening, data extraction, and bias assessment were conducted independently. We pooled data using a Bayesian framework and compared between treatment modalities and individual treatments. We ranked the treatment options (using surface under cumulative ranking curve [SUCRA]) and estimated the absolute median frequency for each outcome. Results We included 96 trials with a combined 52,811 participants, published between 2011 and 2021. The median trial sample size was 559, the median participant age was 62 years, and 37% of the overall trial participants were female. A minority of trials (35.4%) used a double-blind design. The most frequent cancer types were non-small cell lung cancer (28 trials, 17,014 combined participants) and melanoma (15 trials, 8,008 combined participants). The most frequently studied interventions were pembrolizumab alone (15 trials, 9,322 combined participants) and nivolumab alone (14 trials, 5,779 combined participants). Overall, 82% of trials had a high overall risk of bias. All included trials provided data for overall irSAEs. Across all outcomes, the treatment modalities with the lowest rank, and therefore with the highest likelihood of irSAEs, were anti-CTLA-4 agents and ICI combinations (SUCRA for overall irSAEs 7% and 10%, respectively), both with an increased odds of irSAEs versus conventional therapy (OR24.41, 95% CrI 11.11 to 58.04 and 22.46, 95% CrI 11.86 to 44.91, respectively). The lowest- and highest-ranked ICIs (excluding drug combinations) for overall irSAEs were ipilimumab (SUCRA 14%; absolute event rate 848 per 10,000 patients, 95% CI 252 to 2,610) and atezolizumab (SUCRA 82%, absolute event rate 119 per 10,000 patients, 95% CI 54 to 279), respectively. Conclusions The overall risk of developing irSAEs differs across ICIs, as well as the irSAE in question. The differing magnitudes of the increased risks is likely clinically relevant, and therefore should be considered in the management of individual cancer patients.
Objective To estimate the frequency of immune-related serious adverse events (irSAEs) in cancer patients receiving immune checkpoint inhibitors (ICI). Methods We searched CENTRAL, MEDLINE, Embase, and clinicaltrials.gov for randomized controlled trials from inception to February 2021. We included randomized controlled trials investigating any ICI for any type of cancer. The primary outcome was overall irSAEs, and secondary outcomes included specific irSAEs of interest. Screening, data extraction, and bias assessment were conducted independently. We pooled data using a Bayesian framework and compared between treatment modalities and individual treatments. We ranked the treatment options (using surface under cumulative ranking curve [SUCRA]) and estimated the absolute median frequency for each outcome. Results We included 96 trials with a combined 52,811 participants, published between 2011 and 2021. The median trial sample size was 559, the median participant age was 62 years, and 37% of the overall trial participants were female. A minority of trials (35.4%) used a double-blind design. The most frequent cancer types were non-small cell lung cancer (28 trials, 17,014 combined participants) and melanoma (15 trials, 8,008 combined participants). The most frequently studied interventions were pembrolizumab alone (15 trials, 9,322 combined participants) and nivolumab alone (14 trials, 5,779 combined participants). Overall, 82% of trials had a high overall risk of bias. All included trials provided data for overall irSAEs. Across all outcomes, the treatment modalities with the lowest rank, and therefore with the highest likelihood of irSAEs, were anti-CTLA-4 agents and ICI combinations (SUCRA for overall irSAEs 7% and 10%, respectively), both with an increased odds of irSAEs versus conventional therapy (OR24.41, 95% CrI 11.11 to 58.04 and 22.46, 95% CrI 11.86 to 44.91, respectively). The lowest- and highest-ranked ICIs (excluding drug combinations) for overall irSAEs were ipilimumab (SUCRA 14%; absolute event rate 848 per 10,000 patients, 95% CI 252 to 2,610) and atezolizumab (SUCRA 82%, absolute event rate 119 per 10,000 patients, 95% CI 54 to 279), respectively. Conclusions The overall risk of developing irSAEs differs across ICIs, as well as the irSAE in question. The differing magnitudes of the increased risks is likely clinically relevant, and therefore should be considered in the management of individual cancer patients.
Descrição
Trabalho Final do Curso de Mestrado Integrado em Medicina, Faculdade de Medicina, Universidade de Lisboa, 2022
Palavras-chave
Revisão sistemática Meta-análise network Ensaios clínicos Inibidores de checkpoint imunitário Neoplasias