Is cancer so “uncurable” we only hope to extend the patient’s life and not cure it completely?
In: Biology
Cancer research statistician here: some answers here are in the ballpark, some completely missing the mark. An exact answer is complex. In short, those rates or probabilities are given because the studies used to generate them are designed with 3- or 5-year survival in mind.
In many cancers, those endpoints are chosen because we may have data to suggest that if you survive to 3 years, then the probability of surviving long term after that is really good. If your disease is going to respond to a treatment, then it likely will in the first 3-5 years. If you’re alive after that, then outcomes stabilize quite a bit.
Further, we design our clinical trials using those endpoints to be able to compare new treatments to historical data. Those endpoints might be considered the standard set by the FDA as to how to estimate efficacy for a specific cancer. We then take the data from those studies to create prediction models or clinical nomograms which are tools that will allow a doctor to talk through with their patient what is their 3 or 5 year survival probability, something that pretty much everyone can easily digest and interpret.
Ideally, we would want to be able to tell someone their long-term survival chances. But that takes a long time conduct studies around – and clinical research is already a slow process. In some cases 3-5 years isn’t very helpful, and so cancer trials may utilize other surrogate endpoints to try to assess endpoints more quickly (objective response rate, event free survival, progression free survival, etc.) that should correlate with long-term survival.
Latest Answers