Quick note for: Improved prediction of PARP inhibitor response and identification of synergizing agents through use of a novel gene expression signature generation algorithm
My friend has just sent me this paper and I have just finished skim through it. This is the note for myself on paper (it is quite fun to read and get the ideas both from the paper and from my friend); Paper -- Improved prediction of PARP inhibitor response and identification of synergizing agents through use of a novel gene expression signature generation algorithm (doi: 10.1038/s41540-017-0011-6).
The team used two different databases to generate the algorithm (iterative resampling analysis to predict sensitivity -- IRAPS)
1. Cancer cell line encyclopedia (focus on solid tumor cell line) --> contains the genotypic as well as gene expression profile of each cancer cell line
2. Genomics of drug sensitivity --> shows the relationship between the mutational status of cancer cells that are sensitive to particular drugs
After generating the algorithm, they used theirs to test with the available cohort, in this case, cisplatin treatment which lacks a good responsive marker to determine the effectiveness of treatment.
The algorithm is also applied to PARPi treatment (focus on BRCA-mutant ovarian and breast cancer). Based on the evidence which showed that there were some groups of patients could not respond well to the PARPi, this is due to there are another mutations that restoring the HR-pathway. This algorithm could be used to select the patients that will get the most benefits out of PARPs inhibitor. Another application is to increase the sensitivity toward PARPi by predicting the drug inhibitor that accelerates the action of PARPi (BRCA defects-HR restore-less sensitivity toward PARPi --> finding another inhibitor --> restore PARPi sensitivity). Also they mentioned on the BRCAness (defect in HR that showed the phenotype like BRCA1/2 mutation) which this algorithm could help the patients who don't have BRCA1/2 mutation but response well toward PARPi (it was found out later that these group of patients have defect in HR pw and show similar phenotype as BRCA1/2 mutation --> this group might be excluded from PARPi treatment from the beginning based on the current genotypic testing)
They also did the bench works to confirm the predictive algorithm accuracy.
Available databases using in this study (actually there are more)
Cancer cell line encyclopedia
https://portals.broadinstitute.org/ccle
Genomics of Drug Sensitivity in Cancer (GDSC)
http://www.cancerrxgene.org/
Cancer Therapeutics Response Portal (CTRP)
https://portals.broadinstitute.org/ctrp/
The Library of Integrated Network-Based Cellular Signatures (LINCS)
http://www.lincsproject.org/LINCS/
welcome to the big-data analysis to get the precise result and fight with the dynamics of cancer cells!!!
Of note; it reminds me the work of SISP group that they focus on the drug response profiles from different kinds of the patient-derived cell lines. Hopefully, they could come up with a very good algorithm that could improve the treatment, esp, for Thai patients.
The team used two different databases to generate the algorithm (iterative resampling analysis to predict sensitivity -- IRAPS)
1. Cancer cell line encyclopedia (focus on solid tumor cell line) --> contains the genotypic as well as gene expression profile of each cancer cell line
2. Genomics of drug sensitivity --> shows the relationship between the mutational status of cancer cells that are sensitive to particular drugs
After generating the algorithm, they used theirs to test with the available cohort, in this case, cisplatin treatment which lacks a good responsive marker to determine the effectiveness of treatment.
The algorithm is also applied to PARPi treatment (focus on BRCA-mutant ovarian and breast cancer). Based on the evidence which showed that there were some groups of patients could not respond well to the PARPi, this is due to there are another mutations that restoring the HR-pathway. This algorithm could be used to select the patients that will get the most benefits out of PARPs inhibitor. Another application is to increase the sensitivity toward PARPi by predicting the drug inhibitor that accelerates the action of PARPi (BRCA defects-HR restore-less sensitivity toward PARPi --> finding another inhibitor --> restore PARPi sensitivity). Also they mentioned on the BRCAness (defect in HR that showed the phenotype like BRCA1/2 mutation) which this algorithm could help the patients who don't have BRCA1/2 mutation but response well toward PARPi (it was found out later that these group of patients have defect in HR pw and show similar phenotype as BRCA1/2 mutation --> this group might be excluded from PARPi treatment from the beginning based on the current genotypic testing)
They also did the bench works to confirm the predictive algorithm accuracy.
Available databases using in this study (actually there are more)
Cancer cell line encyclopedia
https://portals.broadinstitute.org/ccle
Genomics of Drug Sensitivity in Cancer (GDSC)
http://www.cancerrxgene.org/
Cancer Therapeutics Response Portal (CTRP)
https://portals.broadinstitute.org/ctrp/
The Library of Integrated Network-Based Cellular Signatures (LINCS)
http://www.lincsproject.org/LINCS/
welcome to the big-data analysis to get the precise result and fight with the dynamics of cancer cells!!!
Of note; it reminds me the work of SISP group that they focus on the drug response profiles from different kinds of the patient-derived cell lines. Hopefully, they could come up with a very good algorithm that could improve the treatment, esp, for Thai patients.
Comments
Post a Comment