Medscapers ahoy. I am David Kerr, Professor of Cancer Medicine at University of Oxford and past President of the European Society for Medical Oncology. Today I want to talk about another prognostic index for colon cancer. This is related to a recent publication[1] in Annals of Oncology by Dr. Huijbers and colleagues from the University of Leiden, The Netherlands, and the excellent medical center there, and also with colleagues, including me, from the University of Oxford.
These investigators looked at the contribution of stromal cells to prognosis in colon cancer. It was a nicely conducted, large study of 710 patients who had participated in one of our adjuvant colon cancer trials, hence our involvement. We [at University of Oxford] supplied the biological materials to our colleagues in Leiden, who did all of the analysis. They looked at the percentage of stromal cells and, using simple morphologic criteria that were validated with a reading by 2 pathologists, showed that patients who have a high fraction of stromal cells [have a worse prognosis]: If more than 50% of the cells are stromal compared with [the percentage of] epithelial cancer cells, [the patient will have] a bad prognosis, with a hazard ratio of 2.0 and a P value of .0001. For patients who have high stromal components in their cancers, the 5-year survival rate is around 69% compared with an 83% survival rate for those with a low stromal involvement.
This was a validation study of 710 patients, which followed from initial observations in a couple hundred patients. Therefore, in terms of looking at American Society of Clinical Oncology criteria and how we should report biomarker evidence, the numbers are good, the biostatistics are strong, and multivariate analysis was done. Because this was a validation study, it is a retrospective-prospective trial of a novel prognostic biomarker that is morphologically simple to characterize, pathologically straightforward, and reveals very interesting data.
It does not surprise me, in a way. I must admit that I am becoming much more interested in the interaction between epithelial cancer cells and stromal cells. The stroma, of course, is composed of fibroblasts, infiltrating microphages, lymphocytes, and so on, and the interaction among these in terms of production of cytokines and growth factors clearly can have an enormous impact on the biology of the epithelial cancer cells.
Although I have spent almost a lifetime working with colleagues like Ian Tomlinson, wanting to understand the somatic tumor mutations and changes that drive the behavior of colon cancer, we must not forget environment, context, and the stroma. Next time you see a patient with colon cancer, perhaps ask the pathologist to check whether the patient has high or low levels of stroma within the tumor, and then consider how those with a high stromal component may have a significantly worse prognosis.
EmoticonEmoticon