Smoking Linked to Poorer Prognosis in Colon Cancer

Patients with stage III colon cancer who have a history of smoking are more likely to have worse outcomes, according to a large, randomized phase III trial result published in the Journal of Clinical Oncology. Researchers analyzed the data from the North Central Cancer Treatment Group phase III adjuvant trial N0147 and found that smoking was significantly associated with shorter disease-free survival (DFS) and time to recurrence in these colon cancer patients, despite accounting for various patient and tumor characteristics. The adverse relationship was most pronounced for those patients whose colon cancer was BRAF wild-type or KRAS-mutated.

“There have been a few prior studies suggesting this relationship in colon cancer patients and, of course, the health consequences of smoking in the general population are well known,” said Amanda Phipps, PhD, assistant professor, epidemiology and assistant member of the Fred Hutchinson Cancer Research Center. “We had suspected that this relationship between smoking and colon cancer survival might differ for patients with certain tumor characteristics, but the specific differences we observed were not entirely expected.”

Prior studies had suggested that smoking is associated with an increased risk of any colorectal cancer and BRAF-mutated colorectal cancer specifically. Therefore, the association of BRAF wild-type colorectal cancer tumors with worse survival was surprising, said Phipps. “This may suggest that smoking has an impact on the different stages of tumor development and progression for these two different types of cancer.”

Compared with never-smokers, those who had any history of smoking had a significantly shorter DFS-the 3-year DFS was 74% for never-smokers compared with 70% for ever-smokers (hazard ratio [HR] = 1.21). Smoking status was associated with a shorter DFS in those colorectal cancer patients with wild-type BRAF, but not mutated BRAF (HR = 1.36 and 0.80, respectively).

Smoking was more strongly associated with a poorer DFS in those patients whose colorectal cancer harbors a mutated KRAS gene compared with those who have a wild-type KRAS gene (HR  = 1.5 and 1.09, respectively). However, only the interaction by BRAF mutation (not by KRAS mutation) was statistically significant (P = .03 and P = .07, respectively). The results were comparable when the time to recurrence, rather than the DFS, was analyzed.

“BRAF and KRAS mutations are increasingly tested for in certain clinical settings to help guide treatment decisions and inform prognosis,” said Phipps. “We also know that these particular mutations can drive different pathways of tumor development, so we often see that risk factors for colorectal cancer differ according to BRAF and KRAS mutation status. Our results indicate that these tumor markers are also important in terms of prognostic factors.”

It is clear that BRAF wild-type colorectal tumors are biologically different from BRAF-mutated tumors, and that KRAS status also results in biologically different tumors. However, how the mutation status of the BRAF and/or KRAS genes and smoking history impacts tumor biology and prognosis is not yet understood. Additionally, the components in cigarette smoke that can impact different growth and progression pathways in colorectal tumors are not clear either.

In future studies, Phipps would like to better understand how modifiable lifestyle factors other than smoking can impact survival and progression of different types of colon cancer. In terms of smoking, “there is still a need to better study and understand the impact of continued smoking after a cancer diagnosis and colon cancer survival, and to better understand what compounds in cigarette smoke could be having an impact on development and progression of different types of colon cancer,” said Phipps.