Chemotherapy or Palliative Care: How to Decide?

David J. Kerr, CBE, MD, DSc, FRCP, FMedSci


December 15, 2017

Hello there. I'm David Kerr, professor of cancer medicine from the University of Oxford, in England.

One of the dilemmas that all of us who practice clinically face is deciding whether or not we should give someone chemotherapy. Clearly, we feel pressure from the patient, the family, and so on, but I always teach our young doctors that the sign of a mature oncologist is knowing when not to treat someone. Do we have any tools that allow us to back up what is often an intuitive feeling with a stronger evidence base?

A very nice study by Lindsay Renfro and colleagues[1]—a cast of the distinguished good and great in the world of colorectal cancerology—was recently published in the Journal of Clinical Oncology. These investigators developed a tool that predicts which of our patients who present with advanced colorectal cancer are most at risk for early death. This is a large collaborative trials group, ARCAD, which has a database of almost 23,000 patients at their disposal. They used multivariable logistic regression modeling to determine which factors to use to develop an algorithm that would define those patients most at risk for death within 90 days of diagnosis or presentation to the healthcare system.

It is no surprise that the factors include age, number of metastases, performance status, body mass index (BMI) (the lower the BMI, the greater the likelihood of death), and BRAF status. Other factors, such as high neutrophil count, hemoglobin, and platelet count, can come into this, presenting an inflammatory picture; high levels of bilirubin [could also be included]a host of different factors that could be combined. They turned this [information] into a nomogram, or calculator, that allows the clinician to make a rather strong prediction.

For example, let us look at an elderly chap, a man who is at least in his 70s or early 80s, with a performance status of 2, multiple sites of metastasis, BRAF-mutant tumor, a high neutrophil count, and perhaps a degree of thrombocythemia at presentation. You get the picture. If you put that patient's variables into the nomogram, you will find that he may have a likelihood of dying within 90 days of around 50%, regardless of treatment.

Should we be offering that patient intensive chemotherapy, with all its attendant problems and adverse effects? Or should we be saying that the likelihood that he will benefit from chemotherapy is relatively small and, therefore, we would be better instituting a more symptomatic approach, involving our colleagues from palliative care and so on?

Thanks very much for listening, as always. For now, Medscapers, over and out.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.