LOCATION MATTERS IN COLORECTAL CANCER

Hello, I am David Kerr. I am professor of cancer medicine at the University of Oxford. One of the subjects that we keep returning to, and that I'm terribly interested in, is how we use molecular markers to make choices in the clinic, whether for prognosis or prediction. Again, as you know, I have a huge interest in colorectal cancer.

An interesting study reported recently in Journal of the National Cancer Institute by Loupakis and colleagues^[1] looked at the characteristics of a cohort of patients with previously untreated metastatic colorectal. They looked at three trials in which the patients received chemotherapy plus or minus bevacizumab. The study population numbered almost 2000.

What they were most keen on looking at was the impact of the site of the primary tumor—of geography—on overall survival, response rates, and other outcomes. They defined right- and left-sided tumors as being proximal and distal to the splenic flexure, respectively.

What they showed was that left-sided tumors have a better outcome. In terms of overall survival in the left-sided group was 22.5 months vs 17 months for the group with right-sided tumors. Response rates to chemotherapy with bevacizumab were different too: 44% of left-sided tumors vs 28% for the right-sided tumors. Multivariable modeling showed these findings were independent of BRAF status, whether the tumors had mucinous histology or not.

Isn't that interesting? There's no doubt that the more we look at tumor heterogeneity, the biological behavior, the more we start to see how different pathways are involved in carcinogenesis and can lead to quite different behavior.

What does this new information mean in practical terms? (Large cohort, good group, excellent statistics—quite a compelling study.) At a simple level, in clinical trials of new treatments for advanced disease, perhaps we should consider the sidedness of tumor, right vs left, as one of the stratifying factors. It has a really mechanistic impact on how you design prospective randomized trials of new agents in colorectal cancer.

The second consideration (this is perhaps a bit premature) is whether we should consider this factor as part of information we give to patients—when I sit down with a patient in a clinic with advanced disease, I say, "This is what we can and cannot do." If cure is unlikely, I say so and tell the patient that we're looking to extend life, to improve the quality of life, to buy time. One wonders eventually whether sidedness of the tumor might affect the quality of that information-giving, [in explaining that patients] with right-sided tumors do significantly worse and have a much worse outlook than those with tumors sided on the left. It's an important piece of work.

Sidedness. No complicated molecular biology, no next-generation sequencing, no Nobel Prizes, but a solid piece of work that seems to be potentially clinically impactful.

As always, thanks for listening. We'd welcome any comments or any posts that you may care to make. For the time being, Medscapers, over and out. Thank you.