Hello GLASS investigators, Thanks again for your help. In my study, I want to compare Platinum paired samples that are TMZ +/- radiation naive to those that have received TMZ+/- radiation. To do this, I navigate to the clinical_surgeries table. In a prior thread, @kcjohnson informed me that these samples are "forward looking", meaning that the TmzTreatment column means this occurred **__after** the surgery and sample collection. As as example, suppose we look at patient GLSS-MD-0026: ``` SampleBarcode CaseSource Sex AgeAtDx VitalStatus TmzTreatment TmzCyclesReceived TmzConcurrence RadiotherapyReceived RadiotherapyDoseGy DoseFractionGy RadiationTreatmentDetails GLSS-MD-0026-R1 male 77 dead 1 7 0 0 0 0 NA GLSS-MD-0026-TP male 77 dead 1 11 1 1 60 30 IMRT ``` That would mean this patient got TMZ and radiation **AFTER **the TP sample was extracted. Then, this patient got 7 cycles of TMZ **AFTER** the R1 sample was collected. This means that this table includes no information on whether the TP samples received TMZ / radiation before the first resection, correct? The problem is, for my study, I only want to compare samples that are TMZ/radiation naive to those that are treated. So, I want to ask, can we safely assume all TP samples **__did not receive TMZ with or without radiation BEFORE the first surgery?** Or, if not, is this data stored somewhere? Thanks so much for all your help.