In general, accurate cancer treatment can not do without two key breakthroughs: First, through a large number of cancer patients, genetic abnormalities data analysis, screening drug targets sensitive to drug treatment; the second is through a large number of cancer cells can maintain the characteristics of the body In vitro model analysis to verify the sensitivity of drug treatment targets. The former with the emergence of sequencing technology, has become a reality. The latter with the development of tumor organs, will also be achieved.
Patient-derived organoids (PDOs), a miniature three-dimensional cell culture model derived from the patient's primary tumor, are cultured in the laboratory. Recently, PDOs are increasingly being used as a preclinical model of cancer, yet there is very little data on their ability to predict therapeutic response in the clinic.
Thus, George Vlachogiannis, the first author of the paper and molecular pathology at the London Cancer Institute, and colleagues generated PDOs from patients with metastatic colorectal and gastrointestinal cancers. These patients have previously been included in clinical phase I and phase II trials.
The results showed that phenotypic and genotypic analysis of PDOs showed that they were highly similar to primary tumors. This shows that PDOs can complement existing methods to determine the sensitivity of cancers to improve treatment.
DOI: 10.1126 / science.aao2774
The new "cancer substitute"
The researchers compared patients with PDOs and clinical trials of anti-cancer drugs. Analysis of 71 patients showed that PDOs showed 100% sensitivity, 93% specificity, 88% positive predictive value and 100% negative predictive value.
The researchers also found that organs from the same patient predicted changes in drug sensitivity at different time points. For example, PDO originating from the initially paclitaxel-sensitive gastroesophageal patients is more responsive to paclitaxel than PDO established by the same patient after disease progression or PDO from paclitaxel-refractory patients.
At the same time, PDOs also reflect the response of some patients to treatment and capture intra-tumor heterogeneity.
Previous scientists also tried to predict the patient's response to treatment, including making "cancer avatars" (mostly human cancer mouse models). But getting an answer as soon as possible to guide treatment has been a challenge, as the mouse model typically takes six to eight months to produce results.
Vlachogiannis said that with this "new weapon," results are available in a matter of months and can even be made faster.
Vlachogiannis's team said the next step is to develop in vitro models of the tumor microenvironment and assess how microenvironmental factors affect the response of the organ to various treatments.
The industry alike
The research results have also been praised by many researchers. Hans Clevers, a Utrecht-style organ-based research institute affiliated with the Hubrecht Institute in the Netherlands, said there had previously been evidence that laboratory results could be reproduced on organs. But this is the first time that the author has presented a large number of such cases and has impressive statistical results. I'm sure this will be one of the key papers in this area.
Sam Behjati, a pediatric oncologist at the University of Cambridge's Cambridge Institute of Sanger Institute in Cambridge, UK, said the study shows that technically, patients' organs can be used to screen cancer candidates for treatment.
Professor Charles Swanton, principal clinician at the British Cancer Institute, said the new approach could help us test potential targeted therapies before clinical trials.
Professor David Cunningham at the Royal Marsden NHS Foundation Trust in the UK says this promising study is driving our growth in personalized medicine and ultimately bringing patients smarter, more friendly and more effective treatments method.