Patients with colorectal cancer who survive at least 5 years are increasingly likely to die from causes that are common in the general population, highlighting the importance of screening and lifestyle modification, suggested a large cohort study conducted in California.1 The analysis of more than 139,000 patients with a colorectal cancer diagnosis found that 95% of all deaths from the disease occurred in the first 5 years after diagnosis, investigators reported at the Society of Surgical Oncology’s Annual Cancer Symposium.
Among patients living past that mark, other causes of death surpassed colorectal cancer on relative terms at 8 years (cardiovascular disease) and 10 years (second primaries and neurologic disease), although the specific pattern differed somewhat by age at diagnosis. Reassuringly, the main types of second primary cancers that occurred were much the same as types of cancer seen in the general population, indicating treatment was not inducing new malignancies.
Colorectal cancer deaths plateau over time, and then other causes of death become more important as patients survive longer.— Aaron Lewis, MD
“Our study defines what health problems are affecting colorectal cancer survivors. We also provide a timeline for when they become important,” commented first author Aaron Lewis, MD, a fellow in complex general surgical oncology at the City of Hope in Duarte, California.
“Our data support the American Cancer Society’s recommendations for supportive care in colorectal cancer survivors, including cancer screening programs, dietary and exercise recommendations, smoking cessation, and alcohol reduction,” he maintained.
As screening and treatment have improved, more patients with colorectal cancer are achieving long-term survival, with roughly 1.2 million people in the United States now living with this diagnosis, including about 800,000 who have passed the 5-year survival mark, Dr. Lewis noted, giving some background to the study.
“Cancer survivors have unique health problems: They suffer from the consequences of treatment and also have risk factors associated with other diseases. So survivorship clinics have been developed to meet the unique needs of cancer survivors and also provide individualized care at appropriate times in follow-up,” he said. However, “little is known about the long-term health problems in colorectal cancer survivors.”
For the study, the investigators merged the California Cancer Registry with the Office of Statewide Planning and Development database, which captures inpatient data. They then identified 139,743 patients with colorectal cancer diagnosed between 2000 and 2011 and ascertained causes of death from death codes, comparing non–long-term survivors (who died within 5 years of diagnosis) and long-term survivors (who lived past that landmark).
Median follow-up for the entire cohort was 65 months, and the 5-year overall survival rate was 59.1%, Dr. Lewis reported.
Analysis of the cumulative number of deaths per 1,000 patients during follow-up showed that colorectal cancer ranked first as a cause of death, followed by cardiovascular disease, second primary cancers, neurologic disease, and pulmonary disease. “Colorectal cancer deaths plateau over time, and then other causes of death become more important as patients survive longer,” he noted.
When relative causes of death were -analyzed, colorectal cancer accounted for about 70% of all deaths in the first year postdiagnosis, but the percentage fell steadily over time. Cardiovascular deaths, initially accounting for only 10% of deaths, surpassed colorectal cancer deaths at 8 years, and neurologic and second primary deaths, each accounting for only about 5% initially, did so at 10 years.
“Not unexpectedly, these trends differ by age,” Dr. Lewis pointed out. For example, among patients younger than age 65 at diagnosis, colorectal cancer remained the leading relative cause of death out to 11 years, when second primaries took over that position. On the other hand, among patients aged 80 or older at diagnosis, colorectal cancer was the leading relative cause of death out to only 3 years, at which time cardiovascular disease took the lead.
When the top 10 causes of death throughout follow-up were analyzed, the most common cause was colorectal cancer for the non–long-term survivors, accounting for 65.4% of all deaths. In contrast, it was cardiovascular disease in the long-term survivors, accounting for 24.0% of all deaths, followed by colorectal cancer (23.2%), neurologic causes (13.0%), and second primaries (12.7%)—a pattern similar to that in the general population.
Types of second primaries were comparable for the non–long-term survivors and the long-term survivors. Cancers of the lung and bronchus had the highest prevalence in each group (24.5% and 31.6%, respectively), followed by non-Hodgkin lymphoma and leukemia combined (12.3% and 10.7%), pancreatic cancer (7.4% and 7.6%), and liver cancer (7.3% and 6.9%).
“These are also similar to the trends in the general population,” Dr. Lewis commented. “Interestingly, in the [non–long-term survivors], other [gastrointestinal] cancer is the number five leading cause of death. It’s unclear whether this is because of some coding error or if there is actually a correlation between colorectal cancer and other [gastrointestinal] malignancies.”
City of Hope’s survivorship clinic has not yet incorporated these new data into its management practices, according to Dr. Lewis. “But I would implement them to counsel patients and help them to understand and perhaps give them a little bit more encouragement to change their diet or exercise or to stick with the cancer screening program or stop smoking,” he said. “You should emphasize these things from the time they are diagnosed with colon cancer, and that’s when you can have an impact. But I think knowing the timeline, you can tell patients what to expect.” ■
Disclosure: Dr. Lewis reported no conflicts of interest.
1. Lewis A, Ituarte PH, Melstrom K, et al: Causes of death in long-term survivors of colorectal cancer. 2017 Society of Surgical Oncology Annual Cancer Symposium. Abstract 53. Presented March 17, 2017.
We need to compare causes of death long term vs the population. Where there are differences, we need to ask why and determine if the treatment we give affects long-term survival.— Patrick S. Sullivan, MD
”This was a great study and novel, asking what longer-term...!-->!-->