The paper addresses an "entrenched" problem in oncology: since the discovery of temozolomide and its clinical application, there has been no further improvement in the treatment of the most common and lethal primary brain tumour, glioblastoma. This is too many years ago now and there have been many failed efforts to improve the situation. Although temozolomide increases survival - applied after surgery to remove the tumour and subsequent radiotherapy - this treatment only improves patients' life expectancy by a few months. This is why this result is promising, as it identifies a drug that could kill tumour cells that are resistant to the aforementioned treatment by following a chemical strategy that multiplies the damage to their DNA.  

The paper is very strong on the chemical strategy and the results are clear even in a patient-derived in vivo (PDX) model of this tumour. Caution is called for by the fact that no additional glioblastoma models are tested (glioblastomas are characterised by great plasticity), and the need still to have a pharmacological strategy to enter clinical trials. Glioblastoma is a type of tumour that has already dashed our hopes on other occasions. However, if the results were validated in more glioblastoma models and if the pharmacological strategy could evolve into a possible clinical trial, it would open the door to treating many other tumours, not just brain tumours, where genomic instability is both a necessary point for their evolution and an Achilles' heel.